Comparison of comorbidity incidences between hypokalemia and normokalemia primary aldosteronism patients

Comparison of comorbidity incidences between hypokalemia and normokalemia primary aldosteronism patients Qinghui Zhang 
Department of Hypertension Jie Chen  Xian-xian wu  De-Hui Kong  Xue Liu  Zhi-wei Yang  Min Liu  ( [email protected] )
Department of Hypertension
DOI: https://doi.org/10.21203/rs.3.rs-1764210/v1
License: This work is licensed under a Creative Commons Attribution 4.0 International License.   Read Full License
Abstract Little information is available on apparent differences in comorbidities between hypokalemia and normokalemia in primary aldosteronism (PA) patients. Our study aimed to evaluate the differences regarding comorbidities of PA patients with different serum potassium levels. A total of 629 PA patients were enrolled and divided into hypokalemic and normokalemic types of PA. The general clinical characteristics and related biochemical parameters were collected in all subjects. The incidences of different comorbidities in hypokalemia and normokalemia PA patients were analyzed. Of the 629 patients (49.86 ± 11.66 years, range 18–79), 26.9% had hypokalemic PA. The systolic blood pressures (158.43 ± 2.35 vs. 150.84 ± 21.57 mmHg; P < 0.001), diastolic blood pressures (96.57 ± 15.51 vs. 91.98 ± 15.77 mmHg; P < 0.05) and plasma aldosterone concentration (P < 0.001) were signicantly higher in hypokalemic PA patients than those in normokalemic PA patients. The hypokalemic PA patients had higher urinary microalbumin than normokalemic PA patients (P < 0.05). There was a negative correlation between serum potassium level and level of orthostatic aldosterone concentration and systolic blood pressure, The incidences of obstructive sleep apnea were higher in hypokalemic than those in normokalemic PA patients (P < 0.05). Our data suggest that high prevalence of comorbidities presented in patients with PA. The hypokalemic PA patients had higher incidence of obstructive sleep apnea than the normokalemic PA patients. Therefore, PA's early diagnosis and treatment, especially for the patients with hypokalemic PA, are essential.
Introduction In the past few decades, primary aldosteronism (PA) has been considered rare and is less than 1% of hypertension. With improved diagnostic technology, especially after that the plasma aldosterone to renin activity ratio (ARR) is used as a screening indicator for PA, many PA patients with normal blood potassium were identied and diagnosed. It is reported that the prevalence of PA in patients with grade 1, 2 and 3 hypertension is 1.99%, 8.02% and 13.2%, respectively[1], while the prevalence of PA in patients with refractory hypertension is higher, about 17–23%[2]. Over the last few decades, studies have shown that when compared with essential hypertension, PA increases the risk of cardiovascular events[3, 4], atrial brillation[5, 6], left ventricular hypertrophy[4, 7], stroke[3, 8], heart failure[8, 9], metabolic syndrome and diabetes [10, 11] and kidney diseases (decline in glomerular ltration rate and albuminuria)[12–14]. In addition, compared with similar essential hypertension, PA is associated with a higher risk of death, even after starting medication to control blood pressure and block mineralocorticoid receptors[8]. Overall, PA is a harmful condition that requires early detection and appropriate treatment.
The prevalence of hypokalemia in PA patients has been reported to range from 9–37% [3, 15, 16]. Strong evidence suggests that hypopotassemia is related to a more severe progression of cardiovascular events and kidney injury[17, 18]. However, little information is available on apparent differences in comorbidities between hypokalemia and normokalemia in PA. Thus, we collected the clinical data of patients with PA to compare comorbidities between hypokalemic PA and normokalemic PA.
Page 3/16
Materials And Methods
Subjects A total of 629 unrelated patients with PA were recruited from hospitalized patients in the Department of Hypertension, Henan People's Hospital, from January 1, 2017, to November 31, 2020. Patients with PA were divided into groups of normokalemic PA who never had recorded hypokalemia below 3.5 mmol/L and hypokalemic PA who had past or current hypokalemia or received potassium supplements. The Ethics Committee of Henan Provincial People's Hospital had approved our study. All the participants were well informed and agreed to sign written informed consent forms.
Blood pressure measurement After a 5-minute sitting rest, each participant's blood pressure was measured three times by a trained and certied observer using an electric sphygmomanometer (Omron HBP-1300, Kyoto, Japan). The average of the three measurements was used for statistical analysis. Before taking these measurements, participants were asked to avoid drinking coffee or tea, avoid smoking and exercise for at least 30 minutes.
Diagnostic criteria for PA Screening criteria[19]:The ARR > 30 in the orthostatic position was suspected to be PA. Specic implementation steps: The patient should rest in bed before 23:00 and not get up or sit up in the morning. Fasting blood was collected to check the decubitus position, plasma rennin activity (PRA), angiotensin II and plasma aldosterone concentration (PAC). Then, after standing upright for two hours, we drew blood to detect PRA, angiotensin II, and PAC. The reex immue method measured the ARR values in the decubitus and orthostatic positions. ARR = PAC (pg/dl) / PRA (ng/dl/h) / 10. We correct the potassium levels to normal before the ARR measurement.
Conrmatory testing[19]: saline infusion test: before the test, the patient must rest in bed for 1hour and drip 2L 0.9% sodium chloride solution for 4 hours. The trial began from 8:00 to 9:00 in the morning. Blood pressure and heart rate should be monitored throughout the process, and PRA, PAC, serum cortisol, and serum potassium should be measured respectively before and after infusion. PAC was used to diagnose aldosterone (> 100 pg/dl) and exclude PAC (< 50 pg/dl) after the saline test.
Exclusion criteria Patients with other types of secondary hypertension: renal hypertension (renal parenchymal hypertension, renovascular hypertension), endocrine hypertension (pheochromocytoma, Cushing syndrome), high blood pressure caused by aortic constriction and drugs (such as oral contraceptives, hormones, use certain Chinese medicines).
Double hydrogen pyridine class calcium antagonists, beta-blockers, angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, alpha methyldopa, clonidine and nonsteroidal anti-
Page 4/16
inammatory drugs were discontinued at least two weeks or more; Those stopped diuretics for at least four weeks. During discontinuation of these medications, the patient's blood pressure can be controlled with non-dihydropyridine calcium antagonists and alpha-blockers.
Diagnostic criteria for comorbidities Carotid atherosclerosis was diagnosed by the results of color doppler ultrasound of carotid artery. Cerebral arteriosclerosis was dened by the results of magnetic resonance angiography of head. The diagnosis of cerebral infarction and cerebral hemorrhage was based on the results of strict neurological examination, computed tomography, or magnetic resonance imaging tests, which were veried from hospital records. Coronary heart disease (CHD) was diagnosed by the results of coronary arteriography, a previous myocardial infarction, or surgery or coronary revascularization. Atrial brillation was counted as such when episodes were documented by either conventional 12-lead surface electrocardiogramor. Estimted glomerular ltration rate (eGFR) < 90ml/min/1.73m2 was used to diagnose aldosterone renal insuciency. The diagnosis of diabetes mellitus was dened by oral glucose tolerance test, previous physician diagnosis, or current anti-diabetic medication. The diagnosis of obstructive sleep apnea (OSA) was based on sleep apnea testing.
Biochemical measurements Biochemical indexes of all patients were performed in Henan Provincial People's Hospital. Venous blood samples were collected from all subjects after fasting for 10 hours. The electrolyte, liver and kidney function, blood lipid and blood glucose were measured. PRA, angiotensin II and PAC were measured by radioimmunoassay. Microalbuminuria was also detected.
Image examination: Magnetic resonance imaging (MRI) of head and adrenal, magnetic resonance angiography (MRA) of head and kidney, color doppler ultrasound of carotid artery and electrocardiograph were performed.
Statistical analysis We used the statistical software package SPSS 22.0 (SPSS Inc., Chicago, IL, USA) to analyze the data. Quantitative variables consistent with normal distribution were presented as the means ± standard deviation. Quantitative variables that do not conform to the normal distribution were expressed as medians and quartiles. Qualitative variables are expressed as percentages. The calculator's information is the Chi-square test. The signicance of the difference was determined by the unpaired t-test, Mann- Whitney-Wilcoxon test or Fisher's exact test where appropriate. Multivariable linear regression analyses were used to determine the correlation between potassium levels and the risk factors. Binary logistic regression was used to assess the risk of complications and potassium levels. The odds ratio (OR) and 95% condence interval (CI) were were calculated using binary logistic regression. A two-sided P value < 0.05 was considered statistically signicant.
Results
Page 5/16
Patient's characteristics Of the 629 patients with PA, 295 (46.9%) were male and 334 (53.1%) were female. The average age was 49.86 ± 11.66 years (range18–79 years). The age distribution characteristics showed that 45–60 years old is the most common age, accounting for 50.2%, followed by 30–44 years old, accounting for 27.9% (Fig. 1). The systolic and diastolic blood pressures were 153.27 ± 20.81 mmHg and 93.22 ± 15.15 mmHg for the entire cohort. A total of 169 patients (26.9%) had a hypokalemic variant of PA. Plasma potassium levels were 3.19 ± 0.37 mmol/L in the hypokalemic cohort and 4.03 ± 0.31 mmol/L in the normokalemic cohort.
General characteristics of PA patients with hypokalemia and normokalemia The general characteristics of the PA patients with hypokalemia and normokalemia are summarized in Table 1. The duration of hypertension and the levels of systolic blood pressure, diastolic blood pressure, blood sodium and urinary microalbumin were higher, whereas the percentage of females and the levels of blood chlorine, and blood calcium were lower in hypokalemic PA than in normokalemic PA (P < 0.05 for all).
Page 6/16
Parameter Normokalemic PA Hypokalemic PA t/x2 P
Number (n%) 460 (73.1) 169 (26.9) - -
Male (n%) 199 (43.3) 96 (56.8) 9.104 0.003
Age (year) 50.29 ± 11.64 48.67 ± 11.64 1.542 0.124
BMI (kg/m2) 25.79 ± 3.60 25.78 ± 3.50 0.027 0.978
Duration of hypertension (year) 3.0 (0.5–10) 5.0 (1–10) -2.532 0.012
Systolic blood pressure (mmHg) 150.84 ± 21.57 158.43 ± 21.35 -3.893 1.10E-4
Diastolic blood pressure (mmHg) 91.98 ± 15.77 96.57 ± 15.51 -3.226 0.001
heart rate (per minute) 79.84 ± 11.80 78.10 ± 11.30 1.641 0.101
Blood potassium(mmol/L) 4.03 ± 0.31 3.19 ± 0.37 28.532 2.39E-115
Blood sodium(mmol/L) 141.78 ± 2.40 142.80 ± 2.60 -4.634 4. 0E-6
Blood chlorine(mmol/L) 105.14 ± 2.89 103.72 ± 2.53 5.125 3.97E-7
Blood calcium(mmol/L) 2.25 ± 0.12 2.22 ± 0.14 2.087 0.037
TC (mmol/L) 4.47 ± 0.96 4.34 ± 0.86 1.424 0.155
TG (mmol/L) 1.81 ± 1.06 1.84 ± 1.35 -0.278 0.781
HDL-C (mmol/L) 1.12 ± 0.25 1.11 ± 0.28 0.327 0.744
LDL-C (mmol/L) 2.59 ± 0.77 2.53 ± 0.75 0.798 0.425
Urea nitrogen (mmol/L) 5.23 ± 1.48 5.12 ± 1.86 0.742 0.459
Creatinine (µmol/L) 61.28 ± 18.72 61.89 ± 20.26 -0.337 0.736
Uric acid 327.49 ± 90.71 322.02 ± 94.15 0.645 0.519
Glycated hemoglobin        
Fasting blood glucose (mmol/L) 5.00 ± 1.34 5.09 ± 1.37 -0.709 0.479
Urinary microalbumin(mg/L) 8.00(3.00-22.14) 19.34(6.25–53.14) -5.288 1.23E-7
eGFR(ml/min) 104.94 ± 15.97 107.51 ± 19.66 -1.629 0.104
PA: primary aldosteronism. BMI, body mass index; TC total cholesterol; TG triglyceride; HDL-C high- density lipoprotein cholesterol; LDL-C low-density lipoprotein cholesterol; The value of duration of hypertension and urinary microalbumin were presented as median (quartile), the difference between the two groups was determined by the Wilcoxon-Mann-Whitney test. eGFR: estimted glomerular ltration rate.
Page 7/16
PRA, angiotensin II and PAC The levels of PAC [clinostatic, 150.05 (121.90-196.55) vs. 209.00 (158.45–307.50); orthostatic, 201.45 (151.00-245.63) vs. 281.00 (189.60–381.00), P < 0.001] and ARR [clinostatic, 70.97 (57.01–92.06) vs. 100.75 (76.60-152.25); orthostatic, 67.88 (46.17–96.10) vs. 93.20 (58.28-143.56), P < 0.001] at diagnosis were different between normokalemic and hypokalemic PA, with signicantly higher levels in the hypokalemic PA. However, clinostatic and orthostatic PRA and angiotensin II levels were not different between hypokalemic and normokalemic patients (Table 2).
Table 2 Comparison of PRA, angiotensin II, and PAC concentrations between hypokalemic and normokalemic
forms of PA. Parameter Normokalemic PA Hypokalemic PA t/x2 P
Clinostatic PRA (ng/ml.h) 0.20 (0.20–0.20) 0.20 (0.20–0.20) -1.371 0.171
Orthostatic PRA (ng/ml.h) 0.20 (0.20–0.40) 0.20 (0.20–0.30) 0.544 0.586
Clinostatic angiotensin II (pg/ml)
Orthostatic angiotensin II (pg/ml)
Clinostatic PAC (pg/ml) 150.05(121.90- 196.55)
209.00(158.45– 307.50)
281.00(189.60– 381.00)
Orthostatic ARR 67.88(46.17–96.10) 93.20(58.28-143.56) -5.057 4.26E-7
PA: primary aldosteronism. PRA: plasma rennin activity; PAC: plasma aldosterone concentration.
Relative factors for blood potassium levels Serum potassium level was taken as the dependent variable and the risk factors, such as age, body mass index (BMI), total cholesterol (TC), triglyceride fatty acid (TG), high density lipoprotein cholesterol (HDL- C), low density lipoprotein cholesterin (LDL-C), systolic blood pressure, diastolic blood pressure, heart rate, orthostatic aldosterone, fasting blood glucose, glycosylated hemoglobin, creatinine and urinary microalbumin, were screened by multivariable linear regression analyses. The results showed that systolic blood pressure and orthostatic aldosterone was negatively correlated with serum potassium level (both P < 0.05, Table 3)
Page 8/16
Table 3 Multiple regression analysis of factors affecting blood potassium level
Parameter B Std.error Beta t P
constant 4.048 0.752 - 5.386 5.03E-7
Age 0.005 0.005 0.116 1.056 0.294
BMI 0.015 0.015 0.102 1.035 0.303
TC (mmol/L) 0.222 0.241 0.393 0.922 0.359
TG (mmol/L) -0.072 0.091 -0.153 -0.788 0.433
HDL-C (mmol/L) -0.359 0.358 -0.156 -1.003 0.318
LDL-C (mmol/L) -0.126 0.268 -0.170 -0.470 0.639
Systolic blood pressure -0.006 0.003 -0.246 -2.381 0.019
Diastolic blood pressure 3.18E-4 0.004 0.010 0.088 0.930
heart rate 0.002 0.004 0.054 0.543 0.588
Orthostatic PAC (pg/ml) -0.002 4.28E-4 -0.434 -4.786 6.0E-6
Glycated hemoglobin -0.034 0.068 -0.054 -0.505 0.615
Fasting blood glucose 0.032 0.063 0.057 0.513 0.609
Creatinine 0.003 0.003 0.113 0.887 0.378
Urinary microalbumin -2.63E-4 3.99E-4 -0.083 -0.660 0.511
BMI, body mass index; TC total cholesterol; TG triglyceride; HDL-C high-density lipoprotein cholesterol; LDL-C low-density lipoprotein cholesterol.
Comorbidities In our cohort, patients with PA had a high prevalence of comorbidities, such as carotid atherosclerosis (41.8%), cerebral arteriosclerosis (21.8%), diabetes mellitus (16.9%), cerebral infarction (11.1%), renal insuciency (10.8%) and coronary heart disease (10.3%). The prevalence rates of obstructive sleep apnea (OSA) in patients with hypokalemic PA were signicantly higher than those in patients with normokalemic PA (OR = 2.927, 95%CI = 1.485–5.772, P = 0.002). There was no signicant difference in the prevalence of carotid atherosclerosis, cerebral arteriosclerosis, cerebral infarction, cerebral hemorrhage, atrial brillation, coronary heart disease, renal insuciency and diabetes mellitus between hypokalemic PA and normokalemic PA (P > 0.05 for all, Table 4).
Page 9/16
Table 4 Prevalence of comorbidities in patients with hypokalemic and normokalemic PA [n (%)].
Comorbidities Total PA Normokalemic PA
Hypokalemic
PA
0.951
0.362
Cerebral infarction 70 (11.1) 47 (10.2) 23 (13.6) 1.384(0.812– 2.359)
0.232
Cerebral hemorrhage 9 (1.4) 7 (1.5) 2 (1.2) 0.775(0.159– 3.768)
0.752
Atrial brillation 40 (6.4) 25 (5.4) 15 (8.9) 1.695(0.871– 3.299)
0.121
0.467
0.430
0.278
0.002
Discussion This study presents a detailed assessment of the largest published PA series in a single center[19-21]. The gender ratio of PA patients showed that women took a more signicant proportion than men, which is consistent with the results of previous studies, suggesting that the risk of PA among women is slightly higher than men[22]. In this study, the age of patients at the onset of PA is 30-59 years, and the most common age is 45-59 years, followed by 30-44 years, similar to that reported in other studies[22, 23]. Hypokalemia prevalence among patients diagnosed with PA was reported between 9–37%[3, 15, 16]. In our cohort, 26.9% of patients were diagnosed with hypokalemic PA variants, consistent with the above data. Therefore, hypokalemia is not a necessary condition for screening and diagnosis of PA, and the importance of hypokalemia as a suggestive clinical feature of PA has declined signicantly.
Our study showed that women were more likely to have normokalemia in clinical characteristics. These results may be partly explained by the aldosterone antagonism secondary to progesterone. No signicant difference was found in age between the two groups. However, the timing of the development of hypertension varies signicantly, with higher long-term among patients with hypokalemic PA. This fact is relevant because recovery after adrenal resection is associated with a shorter time to hypertension
Page 10/16
evolution, as Celen et al.[24] reported. Therefore, a positive attitude seeking a specic diagnosis is crucial, especially in patients with hypokalemic PA.
It must be highlighted that the levels of mean systolic blood pressure, diastolic blood pressure and PAC were higher in patients with hypokalemic PA than in normokalemic PA, similarly to the published Spain cohort[25]. However, the median value of PAC was lower in our cohort than that in the reported European study of normokalemic PA[21]. In addition, ARR values were signicantly higher in patients with hypokalemic PA than with normokalemic PA, which was inconsistent with the Spanish research that there was no signicant difference in ARR between groups[25]. The reason for these conicting results may be the differences in sample size.
A growing body of information from longitudinal and retrospective studies convincingly supports that PA patients had a higher incidence of complications in cerebrovascular events, coronary heart disease, arrhythmia, kidney damage, metabolic syndrome and diabetes compared with essential hypertension[4, 9- 12, 18, 19, 21]. Millie et al. reported that the incidence of stroke (12.9% vs. 3.4%), myocardial infarction (4.0% vs. 0.6%) and atrial brillation (7.3% vs. 0.6%) in patients with aldosterone-induced adenoma and bilateral adrenal hyperplasia was signicantly higher than that in patients with essential hypertension[19]. In addition, in the study by Born-Frontsberg et al., the incidence for cerebrovascular disease, including cerebrovascular stenosis, transient ischemic attacks (TIA), Prolonged Reversible Ischemic Neurological Defect (PRIND) and stroke, was 12.8% in PA. The incidence of cerebrovascular complications did not differ between the hypokalemic PA and normokalemic PA[21]. In our study, the incidence of cerebrovascular events, including cerebral infarction and cerebral hemorrhage, was 12.5%. There was no difference in cerebral infarction and cerebral hemorrhage prevalence between hypokalemic and normokalemic PA patients. The high occurrence of cerebrovascular events among PA patients may be explained by the elevated aldosterone concentrations independently from the blood pressure[26].
Elevated aldosterone concentrations harm endothelial dysfunction, independent of blood pressure, and are associated with microvascular inammation and brain and heart muscle[27, 28]. Studies have reported that cardiovascular complications in patients with PA were higher than those with essential hypertension[4, 29]. In a French cohort, patients with PA had a signicantly higher prevalence of coronary artery diseases (OR, 1.9), nonfatal myocardial infarction (OR, 2.6), heart failure (OR, 2.9) and atrial brillation (OR, 5.0) than properly matched controls with essential hypertension. However, cardiovascular complications were not signicantly increased in PA patients with hypokalemia.[9] Catena et al.[30] also showed that cardiovascular complications were higher in PA than in essential hypertension (35% vs. 11%). The ORs of sustained coronary heart disease, cerebrovascular events and arrhythmias were higher at 2.80, 4.36 and 4.93, respectively. In addition, the study by Born-Frontsberg et al. [21] demonstrated a signicant difference in the prevalence of cardiovascular comorbidities between hypokalemic and normokalemic PA (35.2% vs. 20.6%). The predominance of angina pectoris and chronic cardiac dysfunction was notably higher in the hypokalemic variant of PA than in the normokalemic variant. However, no signicant difference in atrial arrhythmia was found between the two groups. Inconsistent with these ndings, in the present study, the rate of cardiovascular comorbidities, including coronary heart
Page 11/16
disease and arrhythmia, was 16.7%, which was lower than that in the previous studies[21, 30]. There was no difference in atrial arrhythmia and coronary heart disease between patients with hypokalemic PA and hypokalemic PA. However, patients with hypokalemic PA showed trend toward higher prevalences of atrial arrhythmia, which may be accounted by low serum potassium level. Another reason for this result may be high aldosterone level in patients with hypokalemic PA. Elevated aldosterone concentration may be a risk factor for arrhythmias through left ventricular hypertrophy or primarily through left atrial brosis or a combination of both.[19] The possible reason for these conicting results between different studies may be the differences in sample size. Therefore, further investigations with larger sample size are needed to conrm our ndings.
Clinical studies have shown that PA is associated with renal complications, reecting the ability of elevated aldosterone levels to cause renal dysfunction. In a large multicenter cross-sectional PAPY study, 24-hour microalbuminuria was signicantly higher in patients with…