The aldosterone antagonist spironolactone was introduced as an antihypertensive agent in 1959. With the recognition that some 10-15% of hypertensive patients have a raised aldosterone : renin ratio (ARR), suggestive of aldosterone excess, there is renewed interest in this agent. It has also been suggested that a high ARR may predict a good response to spironolactone. While spironolactone is now established in the management of congestive heart failure, recently there have been a number of reports of serious adverse effects particularly due to hyperkalaemia. The object of our study was to determine the response to spironolactone in (i) untreated hypertensive patients with no evidence of a secondary cause and (ii) patients receiving chronic therapy for hypertension but where blood pressure had not been adequately controlled despite standard therapy, in most cases with at least three antihypertensive agents. We also wished to determine whether the ARR would predict the response to therapy.

Thirty patients with untreated hypertension (mean 159  ±  3/95  ±  1 mmHg, mean  ±  SEM) and an ambulatory blood pressure monitoring recording > 135/85 mmHg, were randomized to receive spironolactone 50 mg or alternative diuretic bendrofluazide 2.5 mg for 1 month. Blood pressure was measured by an independent observer using an oscillometric device (Omron HEM 705 CP) before and after therapy.

We also undertook a retrospective review of 43 patients (29 male, 14 female) mean age 58  ±  2 years who had received spironolactone due to an inadequate blood pressure response to standard therapy. Aldosterone and renin were measured seated using radio-immunoassay. Blood pressure and biochemistry were recorded for the visit before the commencement of treatment and on the first return visit. The adverse effects of spironolactone were also recorded. Statistical analysis was performed with JMP Version 3.01 (SAS for Windows). Data are expressed as mean  ±  SEM. The change in haemodynamic and biochemical parameters was analyzed with paired Wilcoxon Rank Sums test. Relationship between different parameters was analyzed using non-parametric methods. P < 0.05 was considered statistically significant.

In the newly diagnosed and spironolactone treated patients the systolic blood pressure (SBP) was reduced by 18  ±  3 mmHg and the diastolic blood pressure (DBP) was reduced by 11  ±  1 mmHg, a greater reduction than with bendrofluazide ( p < 0.05). There was a highly significant correlation between log ARR and the fall in SBP (r = 0.69, p < 0.001) and DBP (r = 0.45, p < 0.05) in this group and with the fall in systolic BP (r = 0.37, p < 0.01) in the total population (n = 73). Patients previously on treatment were on 2.8  ±  0.9 medicines and for many patients a fourth medicine had been tried. The mean duration of hypertension was 5  ±  0.7 years. Following spironolactone the mean duration of follow-up was 3.7  ±  0.4 months. The mean dosage of spironolactone was 42 mg and blood pressure reduction was 28  ±  3/13  ±  2 (167  ±  3/95  ±  2 to 139  ±  3/82  ±  2, p < 0.001). There was a significant increase in potassium concentration from 3.7  ±  0.07 to 4.2  ±  0.09 mmol l^-1 ( p < 0.001) with no significant changes in sodium concentrations. Serum creatinine concentration increased from 93  ±  3 to 101  ±  4 µmol l^-1 ( p < 0.01). Hyperkalaemia was recorded in two patients on ACE inhibitors and it responded to a decrease in spironolactone dose or addition of a thiazide. Five male patients (approximately 15%) developed gynaecomastia during chronic treatment.

These data suggest that the response to spironolactone, particularly in untreated hypertensive patients, is related to the ARR. Spironolactone is effective in patients with ‘resistant’ hypertension and is generally well tolerated.