Aldosterone blockers (mineralocorticoid receptor antagonism) and potassium-sparing diuretics

J Clin Hypertens (Greenwich). 2011;13:644–648. ©2011 Wiley Periodicals, Inc.Key Points and Practical Recommendations

• • Mineralocorticoid receptor (MR) antagonists (aldosterone blockers) provide effective antihypertensive treatment, especially in low‐renin and salt‐sensitive forms of hypertension, including resistant hypertension.
• • Newer, more selective MR antagonists (eg, eplerenone) have fewer of the progestational and antiandrogenic effects than spironolactone, enhancing tolerability and potentially improving adherence to therapy.
• • MR antagonists provide an additional benefit in the treatment of heart failure when combined with angiotensin‐converting enzyme inhibitors, digoxin, and loop diuretics.
• • Other potassium‐sparing diuretics (amiloride or triamterene) are generally prescribed for essential hypertension as a fixed‐dose combination with hydrochlorothiazide.
• • The dose range for spironolactone with resistant hypertension is between 25 mg/d and 50 mg/d, and eplerenone is an appropriate alternative if spironolactone is not tolerated because of sexual side effects.
• • In general, the combined use of spironolactone and adequate doses of a thiazide diuretic or a thiazide‐like agent such as chlorthalidone for the treatment of resistant hypertension maximizes efficacy and reduces the risk of spironolactone‐induced hyperkalemia.

Spironolactone, a nonselective mineralocorticoid receptor (MR) antagonist, has been in clinical use for several decades. A selective MR antagonist (eplerenone) has also been developed and is now in clinical use.Before proceeding to review the use of these drugs in clinical practice, we will briefly review how the aldosterone paradigm and current concepts of the mineralocorticoid receptor have changed.     

Outpatient conversion of treatment to potassium-sparing diuretics

Thiazide diuretics frequently cause a decrease in serum potassium levels. In this study, 34 percent of patients taking hydrochlorothiazide had serum potassium levels below 3.5 meq/liter. The response of the serum potassium level was studied after treatment in 56 patients was switched from 50 mg of hydrochlorothiazide daily to either two capsules of hydrochlorothiazide/triampterene (Dyazide), or one tablet of hydrochlorothiazide/amiloride (Moduretic) daily, over nine to 15 months. The 24 patients whose treatment was changed to Dyazide had a rise in serum potassium levels from a mean of 3.56 meq/liter to 4.17 meq/liter in two to three weeks. The 32 patients whose treatment was changed to Moduretic had a rise in serum potassium levels from a mean of 3.76 meq/liter to 4.14 meq/liter in two to three weeks. The resultant rise in potassium levels was stable throughout the follow-up period in both groups. Patient acceptance of this change was excellent.     

The effects of thiazide and thiazide-potassium sparing diuretics on fibrinolytic system parameters.

OBJECTIVE: This study investigates whether the combination of thiazides with an aldosterone antagonist can decrease their negative effects on the fibrinolytic activity. METHODS: Twenty-eight hypertensive patients (20 men, 8 women) visiting our hypertension unit were included in the study. The control group consisted of age- and gender-matched 9 normotensive healthy individuals. The patients in the 1st group (7 men, 2 women, mean age 48.55+/-6.14 years) were given 50 mg hydrochlorothiazide (HCT), whereas patients in the 2nd group (7 men, 2 women, mean age 48+/-6.3 years) received a combination of 50 mg HCT and 5 mg amyloride and the 3rd group (7 men, 3 women, mean age 48.2+/-7.25 years) took 50 mg HCT and 50 mg spironolactone for a period of 2 weeks. The plasminogen activator inhibitor (PAI)-I, tissue plasminogen activator (t-PA) and PAI-I/t-PA ratio were assessed before and after treatment. RESULTS: Treatment with HCT-spironolactone caused an increase in PAI-I (p<0.001) and t-PA ( p<0.001), while no changes were observed in PAI-I/t-PA (P>0.05). In patients treated with HCT-spironolactone, PAI-I increase rate was lower than in those treated with HCT and HCT-amyloride (p<0.001). Hydrochlorothiazide, HCT-amyloride and HCT-spironolactone treatments caused a significant decrease in the baseline blood pressure values (p<0.001). Uric acid levels had increased after treatment with HCT (p<0.01) and HCT-amyloride (p<0.001), but no changes were observed in individuals receiving HCT-spironolactone (p>0.05). CONCLUSION: Thiazides have a negative effect on the endogenous fibrinolytic activity, which is already impaired in the hypertensive patients. Their use in combination with an aldosterone antagonist such as spironolactone can decrease their hypofibrinolytic effects and metabolic side effects.     

Effect of potassium-sparing diuretics on the renin-angiotensin-aldosterone system and potassium retention in heart failure.

The renin-angiotensin-aldosterone system and electrolyte levels in 11 patients with heart failure controlled on digoxin and frusemide were investigated after separate periods of Slow K, spironolactone, and amiloride therapy. When spironolactone or amiloride replaced Slow K, distinct parallel increments in the levels of renin, angiotensin II, and aldosterone resulted. Though plasma potassium was generally higher after spironolactone and amiloride than after Slow K, exchangeable potassium was similar with the three regimens. There was no significant relation between plasma potassium and concurrent exchangeable potassium.     

Antiangiogenic effects of spironolactone and other potassium-sparing diuretics in human umbilical vein endothelial cells and in fibrin gel chambers implanted in rats

OBJECTIVE: Potassium-sparing diuretics have different effects on angiogenesis that may mediate some abilities to treat cardiovascular diseases. The aim of the current study was to compare the effects of spironolactone and an active metabolite, canrenone, or a derivative, eplerenone, and amiloride, a diuretic without affecting mineralocorticoid activity, on the proliferation of human umbilical vein endothelial cells (HUVEC) and on angiogenesis in fibrin gel chambers implanted in rats. MATERIALS AND METHODS: We measured the effects of spironolactone, canrenone, eplerenone, and amiloride on the proliferation of HUVEC in the presence or absence of vascular endothelial growth factor (VEGF) and basic fibroblast growth factor (bFGF). We also examined the effects of these compounds on migration and capillary tube formation by HUVEC. Finally, the effects of the compounds on neovessel formation in vivo were investigated by implanting Wistar rats for 14 days with perforated Plexiglas chambers containing rat fibrin. RESULTS: Spironolactone and amiloride inhibited the proliferation of HUVEC, but canrenone and eplerenone had no effect. The inhibitory effect of spironolactone was not prevented by VEGF or bFGF. Aldosterone had no effect on spironolactone-induced inhibition of HUVEC proliferation. Spironolactone induced a dose-dependent reduction of both cell chemotaxis and capillary tube formation. In fibrin gel chambers, spironolactone and amiloride significantly reduced the numbers of both peripheral and central neovessels. Canrenone and eplerenone, in contrast, had no antiangiogenic effect. CONCLUSION: Spironolactone and amiloride significantly inhibited angiogenesis in vitro and in the fibrin gel chamber in vivo. Spironolactone antiangiogenic effects are unrelated to antimineralocorticoid activity.     

Short-term effects of exogenous angiotensin II on plasma fibrinolytic balance in normal subjects

OBJECTIVE: Numerous studies have provided evidence for a direct functional link between the renin-angiotensin-aldosterone system and the fibrinolytic system. Angiotensin II has been suggested to mediate this interrelationship because this peptide was shown to stimulate plasminogen activator inhibitor-1 (PAI-1) in experimental settings. However, evidence from studies in man regarding effects of Angiotensin II on fibrinolytic function remains controversial. METHODS: In the present study, we have therefore infused Angiotensin II at doses of 1, 3 and 10 ng kg(-1) min(-1), each over 45 min, in 9 healthy volunteer subjects without and with pretreatment with a single dose of the Angiotensin II (type 1) (AT1)-receptor antagonist valsartan (160 mg). RESULTS: Angiotensin II infusion dose-dependently increased plasma Angiotensin II concentrations and systolic/diastolic arterial blood pressure from 121 +/- 2/70 +/- 2 mmHg to 146 +/- 2/97 +/- 1 mmHg (p < 0.001). The maximum increase in blood pressure was completely abolished (118 +/- 3/72 +/- 1 mmHg) when Angiotensin II was infused in volunteers pretreated with valsartan. In spite of the marked hemodynamic changes seen with the Angiotensin II infusion, no effect could be demonstrated on the activity and antigen concentration of PAI-1. Furthermore, pretreatment of the volunteers with valsartan markedly blunted the increase in arterial blood pressure but was not associated with changes in PAI-1. CONCLUSION: In conclusion, in healthy volunteer subjects, short-term infusion of Angiotensin II without and with concomitant administration of an AT1-receptor antagonist has no effect on PAI-1 activities and plasma concentrations.     

Rates of receptor-dependent and -independent low density lipoprotein uptake in the hamster.

By using a constant infusion technique in the hamster, rates of uptake of [14C]sucrose-labeled hamster low density lipoprotein (hamLDL) and methylated hamster LDL (MehamLDL) were directly measured in 15 tissues. From these measurements the magnitude of LDL receptor-dependent and receptor-independent lipoprotein transport was calculated. The whole-animal clearance of hamLDL equaled 547 microliters/hr per 100 g of body weight. LDL clearance per g of tissue was highest in the liver (114 microliters/hr per g), ovary (43), spleen (36), adrenal gland (29), and intestine (24) and was lowest in fat (0.75), brain (0.35), and muscle (0.26). When adjusted for organ weight, the sum of the absolute clearance rates in all of the tissues examined equaled the rate of whole-animal LDL turnover. Liver accounted for 73%, and the jejunum and ileum combined accounted for 7% of whole-animal clearance. The 12 other tissues each accounted for only a minor portion of LDL clearance. Rates of uptake of Me-hamLDL were much less in many tissues and accounted for only 6-12% of the uptake of LDL in the liver, ovary, adrenal gland, lung, and kidney. However, this receptor-independent uptake was quantitatively more important in the intestine (44%) and spleen (72%) and accounted for essentially all LDL uptake in organs such as muscle, skin, and brain. Thus, in the hamster, most LDL is taken up and degraded by the liver. This uptake process is greater than 90% mediated by the LDL receptor and manifests saturation kinetics. Finally, cholestyramine feeding increases receptor-mediated LDL transport in the liver but in no other tissue studied.     

Short term effects of GnRH agonists on plasma fibrinolytic balance in patients with advanced prostate cancer

Background Gonadotropin releasing hormone (GnRH) agonists are the cornerstone of metastatic prostate cancer treatment. Cardiovascular effects of GnRH agonists are unclear. In this study, we investigated the short term effects of GnRH agonists on plasma fibrinolytic parameters in patients with metastatic prostate cancer. Methods Eleven patients (mean age 69.3 ± 6.5) with metastatic prostate cancer and a clinical indication for GnRH agonist therapy were selected. Plasma plasminogen activator inhibitor (PAI-1) antigen (Ag), tissue plasminogen activator (t-PA) Ag and thrombin-activatable fibrinolysis inhibitor (TAFI) activity levels were measured at baseline and at 4 weeks after the first dose of GnRH agonist, Goserelin Acetate (Zoladex^®, subcutaneous administration, 10.8 mg). Results Serum prostate specific antigen (PSA) levels significantly decreased from 36.6 ± 19.3 to 1.1 ± 0.3 ng/ml after Goserelin acetate treatment (P = 0.005). Significant changes occurred in the fibrinolytic parameters. GnRH agonists decreased plasma t-PA Ag levels (16.3 ± 4.9 vs. 12.2 ± 2.8 ng/ml, P = 0.047) and increased PAI-1/t-PA molar ratio (4.8 ± 3.6 vs. 6.6 ± 3.4, P = 0.16), on the other hand, plasma PAI-1 Ag (59.0 ± 48.5 vs. 56.4 ± 30.5 ng/ml, P = 0.8), and TAFI levels (130.6 ± 9.5 vs. 124.2 ± 26.5% activity, P = 0.3) did not change significantly. Conclusion This study provides evidence that GnRH agonists may inhibit fibrinolytic system by decreasing t-PA levels.     

The effects of TNF alpha inhibition on plasma fibrinolytic balance in patients with chronic inflammatory rheumatical disorders

OBJECTIVE: Recent studies support an inflammatory basis for atherosclerosis. Patients with chronic inflammatory rheumatical disorders are at increased risk for cardiovascular events, and this can be partially attributed to the inhibition of fibrinolytic system. TNF a inhibitors such as infliximab are shown to retard the progression of inflammatory arthritides. In this study, we investigated the effects of infliximab on plasma fibrinolytic parameters. METHODS: Thirteen patients (7 female, 6 male; mean age: 44 +/- 11 years) with a clinical indication for infliximab (rheumatoid arthritis (RA) (n = 8), ankylosing spondylitis (AS) (n = 5)) were selected. Plasma plasminogen activator inhibitor (PAI-1), tissue plasminogen activator (t-PA) antigens (Ag) and high sensitive C-reactive protein (hs-CRP) levels were measured during low salt intake at baseline. All patients received infliximab (Remicaide, i.v. infusion, 3 mg/kg). Plasma PAI-1 Ag, t-PA Ag and hs-CRP were measured during low salt intake at the end of 2 weeks. All samples were collected at 9 AM. Antigen levels were determined using a 2-site enzyme-linked immunosorbent assay. RESULTS: Patients experienced significant improvement in disease related activity scores after infliximab treatment. DAS score (for rheumatoid arthritis) and BASDAI index (for ankylosing spondylitis) decreased significantly after treatment (p = 0.01 and p = 0.04 respectively). Infliximab significantly reduced the marker of inflammation (hs-CRP) (8.3 +/- 3.9 vs. 4 +/- 4.1 mg/L, p < 0.01). Plasma PAI-1 antigen (64.7 +/- 26.9 vs. 40 +/- 31.1 ng/ml, p = 0.03) and PAI-1/t-PA ratio (10.8 +/- 5.9 vs. 6.6 +/- 3.8, p = 0.02) were significantly lower after the treatment. In contrast, plasma t-PA levels were unchanged (9.4 +/- 4.4 vs. 9.0 +/- 4.3 ng/ml, p = 0.73). CONCLUSION: This study provides evidence that TNF alpha inhibition with infliximab decreases PAI-1 Ag level and PAI-1/t-PA ratio, and hence activates fibrinolytic system in patients with chronic inflammatory disorders.     

Human blood basophils produce interleukin-13 in response to IgE- receptor-dependent and -independent activation

Interleukin-13 (IL-13) is a recently discovered immunoregulatory cytokine. The cellular sources of IL-13 and the regulation of its expression are largely unknown. Here we show that human basophils produce IL-13 in response to IgE-receptor (IgER) crosslinking, IL-3, IL- 3 plus C5a, but not C5a alone. Human basophils express IL-13 in a restricted manner since, apart from IL-4, no other cytokines encoded on the cytokine gene cluster (IL-3, IL-5, and granulocyte macrophage- colony-stimulating factor [GM-CSF]), are induced. Highest levels of IL- 13 are formed after IgE-independent activation leading to a prolonged secretion of IL-13. The response to IgER-cross-linking is more transient preferentially inducing IL-4, IL-3 is a unique cytokine regulating IL-13 production by human basophils: Among a large number of cytokines tested, only IL-3 is capable of directly inducing IL-13 expression. Furthermore, although some IL-13 is produced in response to C5a in the presence of IL-5, GM-CSF, IGF-1 or IL-1 beta, IL-3 is by far the most effective. IL-13 production was blocked by actinomycin D and cycloheximide and conditions leading to IL-13 release also lead to the induction of IL-13 mRNA. This study supports an important immunoregulatory role of human blood basophils, owing to their capacity to simultaneously express IL-13 and IL-4 in a restricted manner.     

Differing beta-blocking effects of carvedilol and metoprolol

BACKGROUND: Metoprolol is a beta(1)-selective beta-adrenergic antagonist while carvedilol is a non-selective beta-blocker with additional blockades of alpha(1)-adrenoceptors. Administration of metoprolol has been shown to cause up-regulation of beta-adrenoceptor density and to decrease nocturnal melatonin release, whereas carvedilol lacks these typical effects of beta-blocking drugs. AIMS: To compare beta-blocking effects of metoprolol and carvedilol when applied orally in healthy subjects. Methods: We investigated the effects of single oral doses of clinically recommended amounts of metoprolol (50, 100 and 200 mg) and carvedilol (25, 50 and 100 mg) to those of a placebo in a randomised, double-blind, cross-over study in 12 healthy male volunteers. Two hours after oral administration of the drugs heart rate and blood pressure were measured at rest, after 10 min of exercise, and after 15 min of recovery. RESULTS: Metoprolol tended to decrease heart rate during exercise (-21%, -25% and -24%) to a greater extent than carvedilol (-16%, -16% and -18%). At rest, increasing doses of metoprolol caused decreasing heart rates (62, 60 and 58 beats/min) whereas increasing doses of carvedilol caused increasing heart rates (62, 66 and 69 beats/min), 50 and 100 mg carvedilol failed to differ significantly from the placebo (71 beats/min). CONCLUSIONS: We conclude that clinically recommended doses of carvedilol cause a clinically relevant beta-blockade in humans predominantly during exercise where it appears to be slightly (although not significantly) less effective than metoprolol. On the other hand, the effects of carvedilol on heart rate at rest appear rather weak, particularly in subjects with a low sympathetic tone. This might be caused by a reflex increase on sympathetic drive secondary to peripheral vasodilation resulting from the alpha-blocking effects of the drug. These results might be helpful in explaining why carvedilol, in contrast to metoprolol, may fail to cause up-regulation of beta-adrenoceptor density and does not decrease nocturnal melatonin release. This, in turn, may be a reason for the weak side-effects of carvedilol resulting from the beta-blockade. In addition, our data might be of interest in the interpretation of the forthcoming results of the COMET trial, although it has to be emphasised that they were derived from healthy subjects and, therefore, cannot be directly extrapolated to patients with heart failure.     

Anti-angiogenesis effects of aldosterone antagonist diuretics

Angiogenesis requires endothelial cell proliferation and their vascular rearrangement. A report of inhibiting effect of spironolactone on smooth muscle cell proliferation led us to study in vitro the effects of this drug on the endothelial cell proliferation and migration. Spironolactone (10 to 100 microM) and one of its active metabolite, canrenone (10 to 100 microM), are added to human umbilical vein endothelial cells (HUVEC). Their effect on cellular proliferation is evaluated by measuring the amount of the cellular nucleic acids using a fluorometric assay (CyQuant). Cell migration is measured using a multiwell chamber assay (Transwell). In further experiments, we investigated their effect on the capillary-like tube formation in vitro generated by HUVEC seeded in a three-dimensional biological gel (Matrigel). The VEGF (10 ng/mL) and the bFGF (10 ng/mL) were used as mitotic and cell differentiation factors. Effect on cell cycle distribution is investigated by flow cytometry analysis. Spironolactone inhibits HUVEC proliferation but canrenone does not have any significant effect. The growth promoters VEGF or bFGF do not modify inhibiting effect of spironolactone. Spironolactone (50 microM) and canrenone (50 microM) are without effect on cell migration. Capillary-like networks on Matrigel is not modified by spironolactone or canrenone. Spironolactone inhibits progression through S phase of the cell cycle. Spironolactone inhibits the proliferation of the endothelial cells in vitro but shows no effect on their migration and their rearrangement in capillary-like structures. These data should be confirmed in models of angiogenesis in vivo.     

Ozonized autohaemotransfusion and fibrinolytic balance in peripheral arterial occlusive disease.

The acute effects of a major ozonized autohaemotransfusion on blood fibrinolytic capacity were evaluated in 20 subjects affected by peripheral arterial occlusive disease (PAOD). The parameters examined were tissue-type plasminogen activator (t-PA) and plasminogen activator inhibitor type-1 (PAI-1). In subjects not previously submitted to autohaemotransfusion ('unaccustomed' subjects), whether they were PAOD patients or healthy volunteers, the PAI-1/t-PA ratio in the blood samples taken 15 min before the autohaemotransfusion was higher (P < or = 0.05) than at baseline. These changes were independent of the presence of ozone in the autohaemotransfusion blood. Values in both healthy and PAOD-affected individuals were again at baseline 120 min after the end of autohaemotransfusion. In PAOD patients and in healthy subjects previously submitted to several autohaemotransfusions ('accustomed' subjects), the PAI-1/t-PA ratio did not significantly change before, during and after an additional autohaemotransfusion. The results (the increased heart rate and epinephrine and norepinephrine urinary excretion only in non-accustomed subjects) suggest that the acute fibrinolytic imbalance is caused by the apprehensive state produced by the procedure in unaccustomed subjects. Autohaemotransfusion with ozonized blood per se does not significantly influence the fibrinolytic balance.     

Lercanidipine and losartan effects on blood pressure and fibrinolytic parameters

Antihypertensive agents may modulate fibrinolysis in addition to reducing blood pressure. We conducted a randomized trial to assess the effects of lercanidipine and losartan on blood pressure (BP) lowering and three fibrinolytic parameters: plasminogen activator inhibitor-1 (PAI-1), D-dimer, and fibrinogen. All patients enrolled had essential hypertension and underwent a placebo run-in period of 2 weeks before randomization to either lercanidipine tablets 10-20 mg once daily or losartan tablets 50-100 mg once daily. Twenty-six patients completed this study. After 8 weeks of treatment, both groups of patients had significantly reduced systolic (SBP) and diastolic BP (DBP) (SBP, p = 0.034 and 0.050, respectively; DBP, p = 0.018 and 0.034 for lercanidipine and losartan, respectively). Both drugs were well tolerated. Only in the group treated with lercanidipine was PAI-1 concentration significantly reduced (57.1 +/- 4.7 to 43.1 +/- 4.8 ng/mL, p = 0.047). No difference was found with D-dimer and fibrinogen in either group. This study shows that both lercanidipine and losartan are effective antihypertensive drugs in patients with essential hypertension. Lercanidipine may provide additional benefit in fibrinolysis.