Clinical Science and Molecular Medicine (1976) 50,269-276.
Suppression of the renin-aldosterone system in mild essential hypertension
A. M. K H O K H A R , J . D. H. S L A T E R , T. P. J O W E T T
AND
N. N . P A Y N E
The Cobbold Laboratories, Thorn lnsriiltte of CIinical Science, Middlesex Hospital Medical School, London
(Received 15 August 1975)
SummarY 1. Suppression of the renin-aldosterone system by
expansion of the extracellular fluid volume with extra sodium and mineralocorticoid for 6 days was studied in nine young men with very mild essential hypertension and in ten normotensive young men. 2. Plasma renin activity, measured both supine and after 45” head-up tilt, and the renal excretion of aldosterone 18-glucuronide were similar in both groups. However, after expansion of the extracellular fluid volume, hypertensive patients showed much less suppression of both variables. 3. This difference persisted despite matching for an equivalent degree of expansion of the extracellular fluid volume as indexed by the change in body weight. 4. Administration of extra sodium and mineralocorticoid produced a greater proportional fall of renal aldosterone excretion than of plasma renin activity in both groups and this dissociation was signisccantly more marked in the hypertensive group. 5. We suggest that (i) a relative autonomy of the renin-aldosterone system may be relevant to the pathogenesis andlor perpetuation of essentialhypertension and (ii) that the syndrome of low-renin hypertension is unlikely to be associated with ‘mineralocorticoid’excess. Key words : extracellular fluid volume, low-renin hypertension, mineralocorticoid, sodium loading. Correspondence: Dr J. D. H. Slater, The Cobbold Laboratories, Thorn Institute of Clinical Science, Middlesex Hospital Medical School, London, W.l.
Introduction There is now considerableevidence to show that the interplay between the activity of an intrarenal vascular receptor and the sodium flux across the macula densa is a major determinant of the rate of renin release. Raised arterial pressure, by increasing both transmural pressure and sodium excretory capacity, should be associated with a reduced rate of renin release. But the activity of the renin-aldosterone system is normal in most patients with benign essential hypertension (Laragh, Ulick, Januszewicz, Deming, Kelly & Lieberman, 1960; Cope, Harwood & Pearson, 1962; Ledingham, Bull & Laragh, 1967; Fishman, Kuchel, Liddle, Michelakis, Gordon & Chick, 1968; George, Gillespie & Bartter, 1968; Streeten, Schletter, Clift, Stevenson & Ddakos, 1969) when assessed by conventional physiological techniques. But the resolution of conventional techniques may be inadequate. By analogy with other endocrine systems, a defect of hormone secretion is usually best revealed by suppression of the dominant physiological factor influencing secretion. Since the activity of the renin-aldosterone system is closely and inversely related to the plasma volume, we have set out to reveal a primary abnormality of the system in uncomplicated very mild essential hypertension by studying the extent to which it can be suppressed by plasma volume expansion. Another object of the study was to determine the renin-aldosterone relationship when excess of mineralocorticoid is given. The syndrome of ‘lowrenin hypertension’ has been ascribed to hypersecretion of an, as yet, undiscovered mineralocorticoid 269
270
A . M . Khokhar et al.
hormone (Woods, Liddle, Stant, Michelakis & Brill, 1969; Crane, Harris & Johns, 1972). The hallmark of this syndrome is that the rate of aldosterone secretion (or renal excretion) lies within the normal range, and yet plasma renin activity (or concentration) is abnormally low. Therefore, if an excessive rate of endogenous mineralocorticoid secretion is relevant to the problem of essential benign hypertension, this abnormal pattern of the renin-aldosterone relationship should be reproducible by the administration of exogenous mineralocorticoid. Some of the data reported here were presented to the European Society for Clinical Investigation in April 1974. Patients and methods Physiological context
Nine male hypertensive patients, aged 26-45 years, were compared with ten normotensive male volunteers, aged 19-40 years. All were uncomplaining and in good health, objectively. The hypertensive patients were slightly heavier (69.2-96.6 kg, mean 83.7 kg) than the normotensive subjects (63.9-80.2 kg, mean 71.2 kg), although their respective heights were similar (176-185 cm, mean 180 cm, for the normotensive group and 157-192 cm, mean 179 cm, for the hypertensive group). Seven of the nine hypotensive patients had never received any hypotensive therapy, and in the two that had all drugs were discontinued at least 4 weeks before the study was performed. Cardiac and renal function was assessed by chest X-ray, electrocardiography, rapid-sequence pyelography, renography with 311labelled hippuran and the determination of the 24 h endogenous creatinine clearance (on two or more occasions before study). Other investigations included the estimation of the plasma concentrations of sodium, potassium, chloride, bicarbonate, urea and cortisol and the rate of renal excretion of vanillyl mandelic acid. Hyperaldosteronism was excluded by measurement of the rate of renal excretion of aldosterone 18-glucuronideand of plasma renin activity (recumbent and erect), both bcfore and after sodiumloadingwith added mineralocorticoid for 3 days (Biglieri, Slaton, Kronfield & Schamberlan, 1967). Two hypertensive patients with unprovoked hypokalaemia were excluded but otherwise the group consists of a consecutive series. Hypertensive patients, except one, were studied in
hospital during the 7 days of the study. Normotensive people were studied in hospital only on days 1 and 7 of the investigation. This difference of protocol can only serve to minimize the difference between the groups because orthostatic stimulation of renin release must have been smaller rather than larger in the hypertensive patients. Each subject received 173 mmol of sodium (Slow Na, Ciba Laboratories, Horsham, Sussex) and 0.5 mg of 9a-fluorohydrocortisone (Florinef, E. R. Squibb and Sons Ltd, Twickenham, Middlesex) daily for the last 6 days of the study, in addition to his usual diet. The latter did not show substantial differences between the two groups, since the basal rates of renal sodium excretion (157f65, 1 SD, and 179f54, 1 SD, for the normotensive and hypertensive groups respectively) and renal potassium excretion (91 4 16, 1 SD, and 78 f30, 1 SD, respectively) did not differ. The study in no way perturbed the participants' dietary pattern and it is implicit in our assessment of the data that there were, at most, small and physiologically insignificant changes of food intake during the week of study. So far as sodium is concerned,an unrealistic variation in dietary sodium intake as high as 25% would have led to a variation of only about 11% in total sodium intake, and, even then, any difference between the two contrast groups must have been much smaller and hence insignificant. Renal aldosterone excretion was measured every 8 h on the first and the last day, and every 24 h on the intermediate days. Plasma renin activity was estimated erect at 16.00hours and supine at 09.00 hours after overnight bed rest for 8-9 h and also in response to a 45" head-up tilt for 1 h between 08.00 hours and 09-00 hours, both before and after the expansion of the extracellular fluid volume. Systolic and diastolic blood pressure and body weight were recorded twice daily, the former sphygmomanometrically, using Korotkoff's sounds I and IV for the systolic and diastolic pressure respectively, and the latter on stand-up scales accurate to +SO g, after emptying the bladder at 08.0049.00 hours. The nature of the study was explained carefully and accepted fully by each participant. It was approved by the Clinical Investigation Sub-committee of The Middlesex Hospital. Analytical techniques Blood was collected without stasis into pre-cooled heparinized glass vacuum tubes (Vaccutainer 3200
Renin-aldosteronein hypertension
271
Normal
(a1
Hypertension
I
0
I
I
15
I
I
30
45
60
L l
0
I
I
I
15
30
45
I 60
Timehin)
Fro. 1. Plasma renin activity in response to a 45"head-up tilt before (a) and after (b) expansion of the extracellular fluid with extra sodium and gar-fluorohydrocortisone.Vertical bars indicate t- SEM. Values of P are. shown in (b).
LH; Bector Dickenson UK Ltd, Wembley, Middle sex) and the plasma separated immediately at 4°C. It was frozen immediately and stored at -20°C. Urine was collected into plastic containers and an aliquot of each sample was stored at - 20°C. Plasma renin activity was determined by a radioimmunoassay estimation of the rate of generation of angiotensin I over 3 h at pH 5.5, by the method of Menard & Catt (1972), modified only in that dimercaprol was omitted from the mixture of angiotensinase inhibitors. To establish that, under the conditions of this study, plasma r a i n activity parallels plasma renin concentration, increments of MRC Standard Human Renin (batch no. 68/356) (0.5 and 1.0 GU x 10- 7 were added to plasma samples collected before and after expansion of the extracellular fluid volume from one normal and one hypertensive subject. The rate of generation of angiotensin I increased as a linear function of the amount of renin added in each. The change of plasma renin Concentration with volume expansion was similar to that of PRA(') (68% and 63% respectively in the normal subject and 28% and 31% in the hypertensivepatient), thus indicating that the changes of PRA we have observed are not due to alterations of the activity of renin substrate or to changes in the activity of activators or inhibitors of the renin-substrate reaction. Aldosterone 18-glucuronide was determined in aliquots (2 ml) of timed urine samples by the radioimmunoassay method of Jowett, Slater, Piyasena & Ekins (1973). Abbreviation: PRA, plasma renin activity.
The lowest values for the rate of aldosterone excretion and for the PRA observed were well above the realistic detection limit of both assays (0.07 nmol/l and 0.1 pmol h-lml-l respectively). Plasma and urine samples were also analysed for sodium and potassium concentration (flame photometer, model 343, Instrumentation Laboratory Inc., Massachusetts) and creatinine concentration (Autoanalyser, Technicon, Tarytom, New York). Statistical methoak The results are expressed either as absolute figures or as a percentage change from the initial values. Mean values are given with 1 SEM unless otherwise stated. Mean arterial pressure was calculated as the diastolicpressure plus one-third of the pulse pressure. The data are analysed with Student's t-test used for significance.
Resalts Plasma renin activity Before expansion of the extracellular fluid, PRA was similar in both groups, whether recumbent at 09.00 hours or in response to a 45" head-up tilt (Fig. la). PRA rose progressively from a mean value of 1.93 to a maximum mean value of 3.70 pmol h-'1nl-~f0.13, or +la77 pmol h - k l - ' in the reference group and from 1-85to 3.47 pmol h - l ml- t- 0.15 or 1.62 pmol h - 'ml - in the hypertensive group. The maximum rise was achieved at
+
A . M. Khokhar et al.
272
30 min in both the groups. Thereafter the PRA fell in all participants. On the seventh day after the administration of extra sodium and mineralocorticoid the mean PRA at 09.00 hours fell to 0.55 pmol h - 'ml - +0.046 in the reference group but only to 1.40 pmol h-'ml-' f0.15 in the-hypertensive group, a 2.5-fold difference. This differencepersisted in response to head-up tilt (Fig. lb), but the increment of PRA after the tilt was similar in both groups. It rose progressively to a maximum mean value of 1.76 or + 1.21 pmol h -'ml- in the reference group and 2.78 or 1-38 pmol h-'ml-' in the hypertensive group. Again the maximum value was observed at 30 min. The difference between the two groups was highly or very highly significant at each point. PRA was also measured at 16.00 hours when the subject has been ambulant all day. This again shows that, although there was no difference between the groups initially, the administration of extra sodium and mineralocorticoid separated them. Mean PRA in the reference group fell from 359f0.16 to 1%0+ 0.10 pmol h - 'ml - or 55 %, whereas in the hypertensive group it fell from 3.18 f0.17 to 2.18 k0.15 pmol h-lml - or 31 %. The difference between the groups was highly significant (P 0.05). Correspondingly, the value for the mean rate of renal sodium excretion was higher in the hypertensive group (by 52-1 15 mmo1/24 h) on each study day, although, again, only on the fourth C
Bloodpressure (Fig. 5 ) The mild degree of blood pressure elevation in the hypertensive group is indicated by the mean values for the systolic and diastolic pressure on the iirst day, which were 143 k 3 and 93 f 4 mmHg respectively in the hypertensive group, and 113 + 4 and 71 f3 mmHg respectively in the reference group. The values are, however, highly significantly different (t = 5-8, P < 0.001 for the systolic pressure and t = 4-1, P
Suppression of the renin-aldosterone system in mild essential hypertension
A. M. K H O K H A R , J . D. H. S L A T E R , T. P. J O W E T T
AND
N. N . P A Y N E
The Cobbold Laboratories, Thorn lnsriiltte of CIinical Science, Middlesex Hospital Medical School, London
(Received 15 August 1975)
SummarY 1. Suppression of the renin-aldosterone system by
expansion of the extracellular fluid volume with extra sodium and mineralocorticoid for 6 days was studied in nine young men with very mild essential hypertension and in ten normotensive young men. 2. Plasma renin activity, measured both supine and after 45” head-up tilt, and the renal excretion of aldosterone 18-glucuronide were similar in both groups. However, after expansion of the extracellular fluid volume, hypertensive patients showed much less suppression of both variables. 3. This difference persisted despite matching for an equivalent degree of expansion of the extracellular fluid volume as indexed by the change in body weight. 4. Administration of extra sodium and mineralocorticoid produced a greater proportional fall of renal aldosterone excretion than of plasma renin activity in both groups and this dissociation was signisccantly more marked in the hypertensive group. 5. We suggest that (i) a relative autonomy of the renin-aldosterone system may be relevant to the pathogenesis andlor perpetuation of essentialhypertension and (ii) that the syndrome of low-renin hypertension is unlikely to be associated with ‘mineralocorticoid’excess. Key words : extracellular fluid volume, low-renin hypertension, mineralocorticoid, sodium loading. Correspondence: Dr J. D. H. Slater, The Cobbold Laboratories, Thorn Institute of Clinical Science, Middlesex Hospital Medical School, London, W.l.
Introduction There is now considerableevidence to show that the interplay between the activity of an intrarenal vascular receptor and the sodium flux across the macula densa is a major determinant of the rate of renin release. Raised arterial pressure, by increasing both transmural pressure and sodium excretory capacity, should be associated with a reduced rate of renin release. But the activity of the renin-aldosterone system is normal in most patients with benign essential hypertension (Laragh, Ulick, Januszewicz, Deming, Kelly & Lieberman, 1960; Cope, Harwood & Pearson, 1962; Ledingham, Bull & Laragh, 1967; Fishman, Kuchel, Liddle, Michelakis, Gordon & Chick, 1968; George, Gillespie & Bartter, 1968; Streeten, Schletter, Clift, Stevenson & Ddakos, 1969) when assessed by conventional physiological techniques. But the resolution of conventional techniques may be inadequate. By analogy with other endocrine systems, a defect of hormone secretion is usually best revealed by suppression of the dominant physiological factor influencing secretion. Since the activity of the renin-aldosterone system is closely and inversely related to the plasma volume, we have set out to reveal a primary abnormality of the system in uncomplicated very mild essential hypertension by studying the extent to which it can be suppressed by plasma volume expansion. Another object of the study was to determine the renin-aldosterone relationship when excess of mineralocorticoid is given. The syndrome of ‘lowrenin hypertension’ has been ascribed to hypersecretion of an, as yet, undiscovered mineralocorticoid 269
270
A . M . Khokhar et al.
hormone (Woods, Liddle, Stant, Michelakis & Brill, 1969; Crane, Harris & Johns, 1972). The hallmark of this syndrome is that the rate of aldosterone secretion (or renal excretion) lies within the normal range, and yet plasma renin activity (or concentration) is abnormally low. Therefore, if an excessive rate of endogenous mineralocorticoid secretion is relevant to the problem of essential benign hypertension, this abnormal pattern of the renin-aldosterone relationship should be reproducible by the administration of exogenous mineralocorticoid. Some of the data reported here were presented to the European Society for Clinical Investigation in April 1974. Patients and methods Physiological context
Nine male hypertensive patients, aged 26-45 years, were compared with ten normotensive male volunteers, aged 19-40 years. All were uncomplaining and in good health, objectively. The hypertensive patients were slightly heavier (69.2-96.6 kg, mean 83.7 kg) than the normotensive subjects (63.9-80.2 kg, mean 71.2 kg), although their respective heights were similar (176-185 cm, mean 180 cm, for the normotensive group and 157-192 cm, mean 179 cm, for the hypertensive group). Seven of the nine hypotensive patients had never received any hypotensive therapy, and in the two that had all drugs were discontinued at least 4 weeks before the study was performed. Cardiac and renal function was assessed by chest X-ray, electrocardiography, rapid-sequence pyelography, renography with 311labelled hippuran and the determination of the 24 h endogenous creatinine clearance (on two or more occasions before study). Other investigations included the estimation of the plasma concentrations of sodium, potassium, chloride, bicarbonate, urea and cortisol and the rate of renal excretion of vanillyl mandelic acid. Hyperaldosteronism was excluded by measurement of the rate of renal excretion of aldosterone 18-glucuronideand of plasma renin activity (recumbent and erect), both bcfore and after sodiumloadingwith added mineralocorticoid for 3 days (Biglieri, Slaton, Kronfield & Schamberlan, 1967). Two hypertensive patients with unprovoked hypokalaemia were excluded but otherwise the group consists of a consecutive series. Hypertensive patients, except one, were studied in
hospital during the 7 days of the study. Normotensive people were studied in hospital only on days 1 and 7 of the investigation. This difference of protocol can only serve to minimize the difference between the groups because orthostatic stimulation of renin release must have been smaller rather than larger in the hypertensive patients. Each subject received 173 mmol of sodium (Slow Na, Ciba Laboratories, Horsham, Sussex) and 0.5 mg of 9a-fluorohydrocortisone (Florinef, E. R. Squibb and Sons Ltd, Twickenham, Middlesex) daily for the last 6 days of the study, in addition to his usual diet. The latter did not show substantial differences between the two groups, since the basal rates of renal sodium excretion (157f65, 1 SD, and 179f54, 1 SD, for the normotensive and hypertensive groups respectively) and renal potassium excretion (91 4 16, 1 SD, and 78 f30, 1 SD, respectively) did not differ. The study in no way perturbed the participants' dietary pattern and it is implicit in our assessment of the data that there were, at most, small and physiologically insignificant changes of food intake during the week of study. So far as sodium is concerned,an unrealistic variation in dietary sodium intake as high as 25% would have led to a variation of only about 11% in total sodium intake, and, even then, any difference between the two contrast groups must have been much smaller and hence insignificant. Renal aldosterone excretion was measured every 8 h on the first and the last day, and every 24 h on the intermediate days. Plasma renin activity was estimated erect at 16.00hours and supine at 09.00 hours after overnight bed rest for 8-9 h and also in response to a 45" head-up tilt for 1 h between 08.00 hours and 09-00 hours, both before and after the expansion of the extracellular fluid volume. Systolic and diastolic blood pressure and body weight were recorded twice daily, the former sphygmomanometrically, using Korotkoff's sounds I and IV for the systolic and diastolic pressure respectively, and the latter on stand-up scales accurate to +SO g, after emptying the bladder at 08.0049.00 hours. The nature of the study was explained carefully and accepted fully by each participant. It was approved by the Clinical Investigation Sub-committee of The Middlesex Hospital. Analytical techniques Blood was collected without stasis into pre-cooled heparinized glass vacuum tubes (Vaccutainer 3200
Renin-aldosteronein hypertension
271
Normal
(a1
Hypertension
I
0
I
I
15
I
I
30
45
60
L l
0
I
I
I
15
30
45
I 60
Timehin)
Fro. 1. Plasma renin activity in response to a 45"head-up tilt before (a) and after (b) expansion of the extracellular fluid with extra sodium and gar-fluorohydrocortisone.Vertical bars indicate t- SEM. Values of P are. shown in (b).
LH; Bector Dickenson UK Ltd, Wembley, Middle sex) and the plasma separated immediately at 4°C. It was frozen immediately and stored at -20°C. Urine was collected into plastic containers and an aliquot of each sample was stored at - 20°C. Plasma renin activity was determined by a radioimmunoassay estimation of the rate of generation of angiotensin I over 3 h at pH 5.5, by the method of Menard & Catt (1972), modified only in that dimercaprol was omitted from the mixture of angiotensinase inhibitors. To establish that, under the conditions of this study, plasma r a i n activity parallels plasma renin concentration, increments of MRC Standard Human Renin (batch no. 68/356) (0.5 and 1.0 GU x 10- 7 were added to plasma samples collected before and after expansion of the extracellular fluid volume from one normal and one hypertensive subject. The rate of generation of angiotensin I increased as a linear function of the amount of renin added in each. The change of plasma renin Concentration with volume expansion was similar to that of PRA(') (68% and 63% respectively in the normal subject and 28% and 31% in the hypertensivepatient), thus indicating that the changes of PRA we have observed are not due to alterations of the activity of renin substrate or to changes in the activity of activators or inhibitors of the renin-substrate reaction. Aldosterone 18-glucuronide was determined in aliquots (2 ml) of timed urine samples by the radioimmunoassay method of Jowett, Slater, Piyasena & Ekins (1973). Abbreviation: PRA, plasma renin activity.
The lowest values for the rate of aldosterone excretion and for the PRA observed were well above the realistic detection limit of both assays (0.07 nmol/l and 0.1 pmol h-lml-l respectively). Plasma and urine samples were also analysed for sodium and potassium concentration (flame photometer, model 343, Instrumentation Laboratory Inc., Massachusetts) and creatinine concentration (Autoanalyser, Technicon, Tarytom, New York). Statistical methoak The results are expressed either as absolute figures or as a percentage change from the initial values. Mean values are given with 1 SEM unless otherwise stated. Mean arterial pressure was calculated as the diastolicpressure plus one-third of the pulse pressure. The data are analysed with Student's t-test used for significance.
Resalts Plasma renin activity Before expansion of the extracellular fluid, PRA was similar in both groups, whether recumbent at 09.00 hours or in response to a 45" head-up tilt (Fig. la). PRA rose progressively from a mean value of 1.93 to a maximum mean value of 3.70 pmol h-'1nl-~f0.13, or +la77 pmol h - k l - ' in the reference group and from 1-85to 3.47 pmol h - l ml- t- 0.15 or 1.62 pmol h - 'ml - in the hypertensive group. The maximum rise was achieved at
+
A . M. Khokhar et al.
272
30 min in both the groups. Thereafter the PRA fell in all participants. On the seventh day after the administration of extra sodium and mineralocorticoid the mean PRA at 09.00 hours fell to 0.55 pmol h - 'ml - +0.046 in the reference group but only to 1.40 pmol h-'ml-' f0.15 in the-hypertensive group, a 2.5-fold difference. This differencepersisted in response to head-up tilt (Fig. lb), but the increment of PRA after the tilt was similar in both groups. It rose progressively to a maximum mean value of 1.76 or + 1.21 pmol h -'ml- in the reference group and 2.78 or 1-38 pmol h-'ml-' in the hypertensive group. Again the maximum value was observed at 30 min. The difference between the two groups was highly or very highly significant at each point. PRA was also measured at 16.00 hours when the subject has been ambulant all day. This again shows that, although there was no difference between the groups initially, the administration of extra sodium and mineralocorticoid separated them. Mean PRA in the reference group fell from 359f0.16 to 1%0+ 0.10 pmol h - 'ml - or 55 %, whereas in the hypertensive group it fell from 3.18 f0.17 to 2.18 k0.15 pmol h-lml - or 31 %. The difference between the groups was highly significant (P 0.05). Correspondingly, the value for the mean rate of renal sodium excretion was higher in the hypertensive group (by 52-1 15 mmo1/24 h) on each study day, although, again, only on the fourth C
Bloodpressure (Fig. 5 ) The mild degree of blood pressure elevation in the hypertensive group is indicated by the mean values for the systolic and diastolic pressure on the iirst day, which were 143 k 3 and 93 f 4 mmHg respectively in the hypertensive group, and 113 + 4 and 71 f3 mmHg respectively in the reference group. The values are, however, highly significantly different (t = 5-8, P < 0.001 for the systolic pressure and t = 4-1, P
