Predictors of blood pressure response to spironolactone

Resistant hypertension was defined by the Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation and Treatment of High Blood Pressure as ‘the failure to reach goal blood pressure (BP) in patients who are adhering to full doses of an appropriate three-drug regimen that includes a diuretic’ [1]. Subsequently, the American Heart Association introduced the concept of ‘controlled resistant hypertension,’ to indicate those patients who have BP controlled on maximum tolerated doses of at least 4 antihypertensive agents [2]. This modified definition implies that there are patients with ‘resistant’ hypertension who can be controlled by simply adding one or more antihypertensive medications, for example, an antagonist of aldosterone. Therefore, the number of three drugs to define resistant hypertension is arbitrary and it is my opinion that this definition needs to be updated. The incidence and prevalence of resistant hypertension vary widely among different surveys. In the National Health and Nutrition Examination Survey performed in the United States between 2005 and 2008, resistant hypertension was noted in 8.9% of the overall hypertensive population and in 12.8% of adults treated with antihypertensive drugs [3]. Daugherty et al. assessed the incidence of resistant hypertension among ambulatory patients with newly treated hypertension followed in two large integrated health plans. Resistant hypertension was based on the number of hypertension medication prescriptions filled, BP measurements, and adherence data [4]. Among patients with incident hypertension, one in 50 patients (1.9%) met the criteria for resistant hypertension. These patients had a 50% increased risk of cardiovascular and chronic kidney disease compared with nonresistant hypertension. Several factors may contribute to resistance of BP to treatment and it is not our purpose to review all of them. Among those, excessive dietary sodium (Na+) ingestion plays an important role. Pimenta et al.[5] observed a mean reduction in office SBP and diastolic BP of 22.7/9.1 mmHg and a mean reduction in 24-h ambulatory BP of 20.1/9.8 mmHg in response to a low-sodium diet (< 50 mmol/day) compared with a diet greater than 250 mmol/day. These data suggest that patients with resistant hypertension tend to be quite salt-sensitive and strategies to reduce dietary salt intake should be implemented before defining patients as resistant to treatment. Alternatively, adding diuretics other than thiazides may be of benefit. Bobrie et al.[6] introduced the concept of ‘sequential nephron blockade.’ They performed a prospective, randomized, open, blinded, end point study in 167 patients with hypertension resistant to treatment with 300 mg/day irbesartan, 12.5 mg/day hydrochlorothiazide, and 5 mg/day amlodipine. These patients were randomized to sequential nephron blockade (consisting of sequential administration of 25 mg/day spironolactone, followed by 20 and then 40 mg/day furosemide, and 5 mg/day amiloride) or sequential renin–angiotensin system blockade (consisting of 5 mg/day ramipril followed by 10 mg/day ramipril, 5 mg/day bisoprolol, and 10 mg/day bisoprolol). At week 12, the mean between-group difference in daytime ambulatory BP was 10/4 mmHg in favor of sequential nephron blockade. The investigators concluded that in patients with resistant hypertension, sequential nephron blockade induced a large reduction in BP via a progressive increase in sodium depletion, and this intervention was more effective than sequential renin–angiotensin system blockade. The addition of spironolactone to other antihypertensive agents has been found to be particularly useful in patients with resistant hypertension. In the Anglo-Scandinavian Cardiac Outcomes Trial-Blood Pressure Lowering Arm 1411 participants received spironolactone mainly as a fourth-line antihypertensive agent for uncontrolled BP. In this study, SBP and diastolic BP fell 21.9/9.5 mmHg (P < 0.001); BP reduction was largely unaffected by age, sex, smoking, and diabetic status [7]. As not all patients with resistant hypertension respond to spironolactone, several investigators have searched for clinical or biochemical predictors of response. In the study by Chapman et al.[7], the only variables that appeared to weakly predict BP response to spironolactone were lower baseline high-density lipoprotein cholesterol, and lower baseline serum potassium (K+) levels. Baseline serum levels of aldosterone do not seem to predict BP response [8]. In this issue of Journal of Hypertension, Ghazi et al.[9] examined potential predictors of BP response to spironolactone in 79 patients with resistant hypertension. They observed that 24-h urinary sodium (Na+) excretion was a significant predictor of favorable BP response after controlling for potential confounders, such as primary aldosteronism, serum aldosterone levels, and urine or serum K+ levels. The study is of obvious clinical interest, but it falls short from explaining why increased urinary Na+ excretion would be a predictor of BP response to spironolactone. One possibility could be that patients who ingest more salt might accumulate more Na+ and benefit by the administration of an additional diuretic with a different mechanism of action than thiazides. From the study by Ghazi et al. it is impossible to discern whether the benefit is unique to spironolactone or it might be seen with other diuretics. To this regard, a recent study compared the antihypertensive effects of furosemide 40 mg/day and spironolactone 25 mg/day given to two different groups of 15 patients with resistant hypertension. Spironolactone was more effective than furosemide for control of BP [10]. However, in this study, furosemide was administered only once daily even though it is well established that furosemide is short-acting and it should be administered more than once daily to maximize its efficacy. An alternative explanation for the observation reported by Ghazi et al. is that a high-Na+ intake might increase kaliuresis, especially when Na+ reabsorption by the renal cortical collecting tubule is enhanced (as it is in primary hypertension). This could result in a decrease in exchangeable K+ and cellular K+ content (even in the absence of alterations in serum K+), factors that correlate negatively with arterial pressure in primary hypertension [11–13]. Studies have also shown that the BP response to spironolactone in patients with resistant hypertension depends on baseline serum K+ levels. Patients with serum K+ less than 4.0 mml/l manifest a significantly greater decrease in BP than those with baseline serum K+at least 4 mmol/l [14]. In the study by Ghazi et al. serum K+ levels were not different between the two groups of patients examined, but urinary K+ excretion was significantly greater in patients with favorable response to spironolactone than in nonresponders. This would indicate that either responders were eating a diet containing more K+ than nonresponders, or responders were able to maintain a normal serum K+ despite urinary losses of K+. The data provided do not allow this distinction and do not exclude the possibility that responders were K+ depleted. Potassium depletion enhances sodium–hydrogen exchanger type 3 by inducing intracellular acidosis and by stimulating the sympathetic nervous system and the renin–angiotensin system [15,16]. Dietary K+ supplementation has opposite effects. Administration of spironolactone may result in a positive K+ balance and greater BP reduction in patients with K+ depletion related to excessive dietary Na+ intake. Another explanation for the results of this study is that the levels of serum aldosterone or the renal tubular response to aldosterone was greater in responders than in nonresponders. Ghazi et al. claim that the effect of sodium on SBP response was independent of primary aldosteronism, serum and urine aldosterone. However, the levels of serum and urine aldosterone levels tended to be higher and the levels of aldosterone-to-renin ratio and urinary cortisol levels were significantly greater in patients that responded to spironolactone compared with those who did not. In addition, the number of patients with primary aldosteronism was significantly greater among patients who had a favorable response to spironolactone. Levels of aldosterone have to be related to Na+ intake. A higher Na+ intake should be associated with a lower renin and serum aldosterone levels. Indeed PRA levels tended to be lower, whereas serum aldosterone levels tended to be higher, and the aldosterone-to-renin ratio was significantly higher in responders than in nonresponders. All these findings suggest that the secretion of aldosterone in relation to dietary Na+ may have been greater in responders compared with nonresponders. Normotensive patients with a positive family history of hypertension exhibit an inadequate suppression of aldosterone secretion after high salt intake and an exaggerated and prolonged adrenocortical hyperresponsiveness to Ang II [17]. Higher aldosterone values even within the physiologic range may predispose normotensive patients to hypertension trough activation of the sodium–chloride cotransporter in the distal tubule, the epithelial sodium channel in the collecting duct, and the sodium pump, and subsequent sodium retention and potassium loss [18]. Finally, responders were more likely to be female, and to have higher SBP and pulse pressure than nonresponders. An additional limitation of the study is that 56% of patients were African-Americans (this would not explain the difference between responders and nonresponders, but it would make the data not generalizable). ACKNOWLEDGEMENTS Conflicts of interest There are no conflicts of interest.