Download Variations in Individual Organ Release of Noradrenaline Measured

585
Clinical Science (1981) 61,585-590
Variations in individual organ release of noradrenaline
measured by an improved radioenzymatic technique;
limitations of peripheral venous measurements in the
assessment of sympathetic nervous activity
M. J. BROWN, D. A. JENNER, D. J. ALLISON A N D C . T. DOLLERY
Departments of Clinical Pharmacology and Radiology, Royal Postgraduate Medical School, London
(Received 28 November 198016 March 1981; accepted 11 May 1981)
Summary
1. The validity of plasma noradrenaline as an
index of sympathetic nervous activity was
assessed by estimating variation in individual
organ contribution to circulating concentrations.
2. Arteriovenous (A-V) differences in noradrenaline and adrenaline concentration were
measured across several organs in nine patients
with mild essential hypertension, in five with
renal artery stenosis and 15 phaeochromocytoma patients.
3. In patients with phaeochromocytomas the
percentage extraction of noradrenaline and adrenaline (estimated from the A-V differences)
was similar across all organs, suggesting that
adrenaline extraction could be used as a marker
for noradrenaline extraction.
4. In the non-tumour patients the A-V difference for noradrenaline was less than that for
adrenaline across most organs studied, reflecting
the net result of noradrenaline release and
extraction. The estimated contribution of various
organs to the noradrenaline concentrations in
their venous effluent was: heart. 21%; kidney
47%; legs 68%.
5. This pattern of A-V difference proved a
positive diagnostic feature for non-tumour
patients since it was not found even in the
patients with small phaeochromocytomas, whose
peripheral venous noradrenaline concentration
alone did not distinguish them.
Correspondence: Dr M. J. Brown, Department of
Clinical Pharmacology, Royal Postgraduate Medical
School, Du Cane Road, London W12 OHS.
0143-5221/81/110585-06%01.50/1/1
6. The venous-arterial difference across the
adrenal. glands of non-tumour patients was more
than 10-fold greater for adrenaline than that for
noradrenaline. Since the mean arterial concentration of noradrenaline was more than fivefold
higher than that of adrenaline, the normal adrenal
contribution to circulating noradrenaline is likely
to be less than 2%.
7. In the patients with renal artery stenosis
renal venous concentrations of noradrenaline
(from the ischaemic kidney) were higher than
arterial values, but mean arterial values were no
higher than in the essential hypertensive patients.
8. Local variations in sympathetic activity
may occur without altering the plasma noradrenaline concentration measured in peripheral
plasma.
Key words: adrenal medulla, heart, kidney,
noradrenaline, sympathetic activity.
Introduction
The measurement of plasma noradrenaline concentration remains the most convenient tool for
estimating short-term changes in sympathetic
nervous activity in man [ 1, 21. Several unproven
assumptions are, however, involved in the use of
such an index. One of these is that an increased
peripheral vein concentration of noradrenaline
reflects a general increase in noradrenaline release
throughout the body or, conversely, that such a
general increase will necessarily cause a rise in
peripheral venous noradrenaline concentration. A
0 1 9 8 1 The Biochemical Society and the Medical Research Society
586
M . J. Brown et al.
further assumption is that the escape of noradrenaline from synaptic clefts into the bloodstream is
uniform in different organs. An opportunity to
test these assumptions was provided in the
present study by the measurement of the arteriovenous (A-V) difference of noradrenaline across
several organs in patients undergoing catheter
studies in their clinical investigation. Since these
differences are the net result of extraction and
release, the use of adrenaline differences as a
marker for noradrenaline extraction was investigated. The A-V difference for adrenaline
across all organs except the adrenal medulla
reflects extraction alone, since only the adrenal
(outside the brain) can synthesize adrenaline [ 31.
Since adrenaline is cleared by the same routes as
noradrenaline [41, its extraction by a tissue is
likely to be similar to that of noradrenaline. So
the adrenaline difference could be used to predict
that of noradrenaline if it was extracted but not
released; the discrepancy between the predicted
and observed A-V difference for noradrenaline
would then be due to noradrenaline release by the
tissue. The validity of this model was tested in
patients with large phaeochromocytomas, in
whom virtually all circulating catecholamines
were derived from one site, with little contribution from other organs.
Subjects and methods
Studies were performed in patients being investigated for suspected phaeochromocytomas
(24), renal artery stenosis (five) or coronary
artery disease (four). None of the first or third
group had received any medication for at least 2
weeks before investigations. All of the second
group were receiving hydralazine, a /?-adrenoceptor antagonist (propranolol or atenolol) and a
diuretic. Studies were performed 1 h after
premedication with pethidine (50 mg) and promazine (50 mg); general anaesthesia was not
used. In the patients with elevated plasma
catecholamine concentrations selective venous
sampling was performed to determine whether
there was a single (tumour) source for the
elevated secretion and, if so, to localize it [51. In
13 patients a tumour was identified (and subsequently removed) in one or both adrenals, and
in two others an extra-adrenal phaeochromocytoma was found. No tumour was found in the
other nine patients (of this group) and introduction
of the pentolinium suppression test [61 subsequently enabled the diagnosis of phaeochromocytoma to be excluded; these patients all had mild
hypertension [blood pressure (BP) 140/95-160/
1101 and their vanillylmandelic acid excretion
and plasma noradrenaline concentrations were
elevated on at least two occasions up to two-fold
above normal; urinalysis, creatinine clearance,
ECG, chest X-ray and intravenous pyelogram
were all normal.
In all the suspected phaeochromocytoma
patients blood samples (5 ml) were obtained from
the adrenal, renal, hepatic, iliac (internal and
external), antecubital and brachial veins, from
several levels in the inferior and superior vena
cava, and from the right atrium. A catheter was
inserted by the Seldinger technique under local
anaesthesia into the right femoral vein and
introduced into most of these sites, under
fluoroscopic control; its position was ascertained
by the hand-injection of 2-3 ml of Hypaque. Since
the left adrenal drains into the renal vein, blood
was drawn from the bifurcation of the renal vein
into upper and lower pole veins; adrenal vein
contamination was excluded by measurement of
cortisol concentration in the renal vein and
arterial samples [ 6 , 71. The arm blood samples
were taken by direct venepuncture, the brachial
vein being located just medial to the artery.
Arterial samples were obtained simultaneously
with the venous samples from a catheter in the
femoral artery.
Five patients with renal artery stenosis (angiographically proven on a subsequent occasion)
were investigated during the measurement of split
renal vein renin concentrations. The patients all
had severe hypertension not controlled by triple
therapy (BP 150/100-175/115). Most had ECG
evidence of left ventricular hypertrophy and mild
impairment of renal function (creatinine
clearance 46-78 ml/min). Blood samples were
obtained from both renal veins, from the inferior
vena cava above and below the orifices of these
veins (‘high’ and ‘low’ inferior vena cava), and
an arterial sample was drawn from the femoral
artery at the end of the study. Adrenal contamination of the renal vein samples was excluded as before. Simultaneous coronary sinus
and left ventricular blood samples were obtained
from four patients undergoing cardiac catheterization in the investigation of angina who
subsequently proved to have normal coronary
artery angiograms.
Informed consent was obtained from each
patient and the research aspects of the investigations were approved by the Ethics Committee of the Royal Postgraduate Medical School.
Analytical method
Blood for catecholamine estimation was collected into chilled lithium/heparin tubes and the
Measurement of noradrenaline release
587
plasma separated within 30 min at 4OC. This was
stored at -8OOC and assayed by a doubleisotope enzymatic technique t81.
Results
In the suspected phaeochromocytoma group 15
patients were found to have a tumour, of whom
eight had peripheral noradrenaline concentrations more than 10-fold above normal. Both
in these ‘large phaeochromocytoma’ patients and
in the remaining seven ‘small phaeochromocytoma’ patients the A-V differences (Fig. 1)
demonstrated a similar extraction of both adrenaline and noradrenaline by all organs studied:
78 k 9 and 83 f 10% respectively, by the liver;
58 k 8 and 54 f. 6% by the kidneys; 68 f 7 and
68 f. 6% by the legs; 10 f. 9 and 12 f 9% by the
superficial forearm tissue (estimated from arterial
and antecubital vein concentrations); 47 k 9 and
40 5 10% by deep forearm tissue (estimated
from brachial vein concentrations). The pattern
was similar for patients with adrenal or extraadrenal tumours, although the plasma adrenaline
concentration in the latter was not elevated.
Interestingly, it was possible to detect extraction
by the normal adrenal gland in patients with large
tumours in the opposite gland, the absence of a
positive A-V difference being indicative of a
small tumour present in the otherwise apparently
normal adrenal. In all these patients there was a
marked and similar step-up in the noradrenaline
and adrenaline concentrations in the inferior vena
cava from below to above the level of the renal
and adrenal veins, of 253 k 31 and 289 f. 37%
respectively.
In the nine patients of this group in whom no
tumour was found, a very different pattern of
A-V differences was observed (Fig. 2). Net
hepatic extraction of both catecholamines was
not significantly different from that in the
phaeochromocytoma patients. Across all other
organs studied, this was true only of the
adrenaline difference; the A/V difference for
noradrenaline was significantly lower or, in some
cases, reversed. On the assumption that the A-V
difference of adrenaline was a valid marker of
noradrenaline, as well as adrenaline, extraction,
the percentage contribution of each organ to its
venous noradrenaline concentration was calculated; the calculation and the results are shown
in Table 1. Only across the adrenal gland was the
venous concentration of either catecholamine
higher than the arterial value in all patients. The
mean venous-arterial difference for adrenaline
was 48.4 f 5.2 nmol/l, which was more than
10-fold higher than that of noradrenaline (4.6 k
L
2
FIG. 1. Arteriovenous differences (concentration
means k SD) of noradrenaline (El) and adrenaline (E%$
across several tissues in 15 patients with large (a) and
small (b) phaeochromocytomas. The mean arterial
concentrations of adrenaline (----) and noradrenaline (-...) are shown; SD of the noradrenaline values
was k 12 nmol/l in (a) and f 1.7 nmol/l in (b); for
adrenaline the SD was k 4 nmol/l in (a) and f 0.21
nmol/l in (b).
T
FIG. 2. Arteriovenous differences of noradrenaline
(0) and adrenaline (B) across several tissues in nine
non-tumour patients. The mean arterial concenand noradrendme
trations of adrenaline (----)
(. . . .) are shown; their SD were respectively i 0.17
and f 1.4 nmol/l.
0.28 nmol/l). Since, by contrast, the peripheral
(arterial) noradrenaline concentration was more
than fivefold that of adrenaline, the contribution
of the adrenal medulla to circulating noradrenaline may be estimated to be less than 2% (1/10
x 1/5), assuming that both catecholamines have
a similar clearance from plasma. In these
non-tumour patients there was no significant
step-up in noradrenaline concentration from ‘low’
to ‘high’ inferior vena cava, although that for
adrenaline was similar to the increase observed in
the phaeochromocytoma patients (264 f 30%).
In the renal artery stenosis patients the renal
venous concentration of noradrenaline was higher than arterial on the ischaemic side, but lower
than arterial contralaterally (Fig. 3). In all these
M . J . Brown et al.
588
TABLE1. Estimaled contribution oj various tissues to the noradrenaline in their venous effluent
This was calculated for each tissue (n = number of patients) on the assurnptiori that the percentage
extraction of noradrenaline (NA) and adrenaline (AD) were the same, but only N A was simultaneously released. The equation is:
venous N A concn. venous AD concn
x 100
contribution (%) =
arterial N A concn. arterial AD concn.
~
~
Patients (diagnosis)
Tissue
Mild essential hypertension (11
Angina (n = 4)
Renal artery stenosis (11
=
9)
5)
36 f 7.4
I 2 f 11
,
Kidney
(ischaemic)
(contralateral)
FIG. 3. Arteriovenous differences of noradrenaline
(0)and adrenaline (B)across the kidneys in five
patients with renal artery stenosis. The mean arterial
concentrations of adrenaline (----) and noradrenaline (. . . .) are shown; their SD were respectively ?
0.25 and 1.6 nmol/l.
7.5-
-::
3.7 5 . 2
47 f 6.1
68 f 9.3
Liver
Kidneys
Legs
Forearm
(deep)
(superficial)
Heart
=
Contribution (means f SD)(%)
L.
I
C
6
FIG. 4. Arteriovenous differences of noradrenaline
(0)and adrenaline (@I
across the heart in four
patients under investigation for angina. The mean
arterial concentrations of adrenaline (----)
and
noradrenaline (....) are shown; their SD were respectively 0.96 and & 2.1 nmol/l.
patients there was a step-up in noradrenaline
concentration from 'low' to 'high' inferior vena
cava ranging from 16 to 55%. The percentage
contribution of the two kidneys to their venous
noradrenaline concentrations, calculated from the
21 f 5.2
81k 11.6
39 f 6.7
A-V differences of noradrenaline as before, was
different on the two sides (Table 1). The cardiac
contribution to the noradrenaline concentration
in its venous effluent was estimated in four
patients (Fig. 4, Table 1). The calculation is
based on the same assumption as before,
although in this instance no comparable data
were available from the phaeochromocytoma
patients to confirm that similar extraction of both
catecholamines usually occurs.
Discussion
In most subjects the adrenal gland was the only
site where the venous noradrenaline concentration was higher than the arterial value,
although it was estimated to make little contribution to circulating noradrenaline. The study
illustrates how, by contrast, other organs may
make an appreciable contribution without any
venous-arterial difference in noradrenaline concentration being detectable. In previous studies
where A-V differences have been measured, it
was not possible to estimate the relative balance
of secretion and extraction, although such investigations have been performed in experimental
animals [91 as well as man [lo, 111. This
difficulty has now been overcome by the simultaneous measurement of noradrenaline and adrenaline differences and by using the percentage
extraction of adrenaline as an estimate of
noradrenaline extraction. The validity of this
estimation depended on (a) showing that the two
catecholamines have similar extractions and (b)
improved catechol 0-methyltransferase assay
precision; since the calculation of noradrenaline
production requires the rimultaneous measure-
Measurement of noradrenaline release
ment of arterial and venous concentrations of
both noradrenaline and adrenaline, any assay
imprecision may be compounded fourfold.
The assumption of similar extraction (of
noradrenaline and adrenaline) is apparently
justified by the data from the phaeochromocytoma patients. In the eight ‘large phaeochromocytoma’ patients, in whom all circulating
noradrenaline could be assumed to be derived
from one source, there was no significant difference in the percentage extraction of noradrenaline and adrenaline (calculated from their
A-V differences) across any organ. The data
from the ‘small phaeochromocytoma’ patients
has been presented separately since their plasma
noradrenaline concentrations were not significantly different from those of the non-tumour group
and may have received some contribution from
sympathetic nerve endings. That this is unlikely,
however, is suggested by their noradrenaline
concentration gradient in the inferior vena cava,
which was no less than that in the ‘large
phaeochromocytoma’ patients, whereas no inferior vena cava gradient was observed in the
non-tumour patients in whom little circulating
noradrenaline was derived from the adrenal
medulla. It has also been shown recently that a
reduction in sympathetic activity by ganglionic
blockade does not reduce plasma noradrenaline
concentration in phaeochromocytoma patients,
even when this is within the normal range [61. If,
then, it is valid to include the data from the ‘small
phaeochromocytoma’ patients, it appears that the
similar extraction of noradrenaline and adrenaline occurs independently of their plasma
concentration.
Although both catecholamines are substrates
for the neuronal and extraneuronal uptake
processes and for metabolism by catechol 0methyltransferase and monoamine oxidase, there
are no assumptions implicit in the present model
regarding the relevant importance of these
pathways in the tissue extraction of the
catecholamines. None will be saturated, since
even the lowest K, (that for neuronal uptake) is
in the micromolar range and higher than the
plasma concentration of even the ‘large
phaeochromocytoma’ patients [41. Similarly,
none of the patients had plasma catecholamine
concentrations approaching the apparent
threshold for extraneuronal accumulation. Nor is
the model inconsistent with the variation in
plasma catecholamine clearance with plasma
concentration [121; this is probably a padrenoceptor-mediated effect, more pronounced
with adrenaline, and most likely due to an
increased cardiac output rather than a specific
589
effect on any of the kinetic processes 1131. No
variation in the relative extraction of the two
catecholamines would be anticipated; this is
supported by the ‘large phaeochromocytoma’
patients with extra-adrenal tumours, in whom a
normal plasma adrenaline concentration was not
associated with any difference in the pattern of
extraction.
The value of the model lies in its ability to
obtain physiological information from purely
clinical data, since such invasive studies could not
readily be performed on normal volunteer subjects. The arguments for its validity are indirect
admittedly, but have spared the need for administering labelled catecholamines to the subjects to measure directly their extraction by each
organ. Clinically, it is of some interest that all
phaeochromocytoma patients proved to have the
same pattern of A-V differences, different from
that of non-tumour patients; the latter pattern,
comprising slight (or absent) A-V difference of
noradrenaline across the legs and kidneys, provides a positive diagnostic feature showing that
the circulating noradrenaline is derived from
sympathetic nerves rather than a tumour [141.
Further studies with measurement of organ
blood flow and cardiac output would be required
to assess the relative contribution of each organ
to circulating noradrenaline. It is possible,
however, to estimate these approximately from
the present data, by using tables of normal values
[ 151, and it is immediately obvious that density of
sympathetic innervation of a tissue correlates
poorly with its contribution to plasma noradrenaline. This may be calculated, for each organ,
from the product of its contribution to its venous
noradrenaline concentration (Table 1) and its
blood flow as a fraction of cardiac output. The
estimated results for heart and legs are 3 and
11% respectively: the reverse of the relative
density of their sympathetic innervation [16, 171.
By a similar calculation the kidneys contribute
12% of circulating noradrenaline.
That cardiac sympathetic nervous activity may
have little effect on peripheral plasma noradrenaline concentrations has been previously suggested [18, 191. Nevertheless, this has not been
widely agreed, and the heart has been considered
to be responsible for the increased plasma
noradrenaline concentration reported in some
patients with essential hypertension [20, 211. If
sympathetic nervous activity varied uniformly in
all organs, it would not necessarily affect the
interpretation of peripheral plasma noradrenaline concentrations if some organs made little
contribution to these (relative to the degree of
sympathetic activity in these organs). But the
590
M . J. Brown et al.
data from the renal artery stenosis patients
suggest that sympathetic activity to an organ
may vary independently of that elsewhere (and be
undetectable by peripheral measurements of
plasma noradrenaline). Although the increased
contribution of the ischaemic kidney to its venous
noradrenaline concentration may partly be due to
reduced flow, the marked step-up in the concentration of noradrenaline from ‘low’ to ‘high’
inferior vena cava suggests a significant abnormality of renal sympathetic activity; yet the
peripheral plasma noradrenaline concentration
was not elevated.
Finally, the relation between peripheral venous
and arterial noradrenaline concentrations are
variable. In the phaeochromocytoma patients all
forearm venous concentrations of both catecholamines were lower than arterial; this was especially true of deep forearm (brachial vein)
levels. In the non-tumour patients this was true
only for adrenaline, suggesting a local contribution to forearm venous concentrations of
noradrenaline. The greater extraction of catecholamines by deep than superficial forearm tissue
may not affect the interpretation of endogenous
noradrenaline concentrations. But investigations
of noradrenaline clearance with infused noradrenaline should measure changes in either arterial
or superficial venous noradrenaline concentration to avoid the risk of measuring local
variations in clearance.
References
111 CRYER,P. E. (1976) Isotopederivative measurements of
plasma norepinephrine and epinephrine in man. Diabetes, 25,
1071-1085.
I21 LAKE, C.R., ZIEGLER, M.G. & KOPIN, I.J. (1976) Use of
plasma norepinephrine for evaluation of sympathetic neuronal
function in man. L f e Sciences, 18, 1315-1325.
131 AXELROD,J. (1962) Purification and properties of phenylethanolamine-N-methyltransferase.Journal of Biological
Chemistry, 237,1657-1660.
I41 IVERSEN,L.L. (1967) The Uptake and Storage of Noradrenaline in Sympathetic Nerves, pp. 114-121. Cambridge
University Press.
151 JONES,D.H., ALLISON,D.J., HAMILTON,
C.A. & REID, J.L.
(1979) Selective venous sampling in the diagnosis of
phaeochromocytoma. Clinical Endocrinology, 10,179-186.
161 BROWN,M.J., ALLISON,D.J., JENNER,D.A., LEWIS, P.J. &
DOLLERY, C.T. (1981) Increased sens
phaeochromocytoma diagnosis achieved by use of plasma
adrenaline estimation and a pentolinium suppression
test. Lancet, I, 174-177.
I71 MANHEM,P., LECEROF, H. & HOKFELT,B. (1978) Plasma
catecholamine levels in the coronary sinus of healthy males at
rest and during exercise. Acta Physiologica Scandinavica, 197,
364-369.
181 BROWN,M.J. & JENNER,D.A. (1981) Novel double-isotope
technique for enzymatic assay of catecholamines, permitting
high precision, sensitivity and plasma sample capacity. Clinical
Science, 61,59 1-598.
I91 ROIZEN,M.F., WEISE, V., Moss, J. & KOPIN, I.J. (1975)
Plasma catecholamines: arterial-venous differences and the
influence of body temperature. Lye Sciences, 16,1133-1 143.
1101 VECHT,R.J., GRAHAM,G.W. & SEVER,P.S. (1978) Plasma
noradrenaline concentrations during isometric exercise. British
Heart Journal, 40,1216-1220.
[Ill WATSON,R.D.S., JONES,D.H., REID, J.L., PAGE,A.J.F. &
LITTLE,W.A. (1979) Plasma noradrenaline concentrations at
different vascular sites at rest and during isometric and
dynamic exercise. Clinical Science, 57, 1 9 ~ .
[I21 CLUITER, W.E., BIER, D.M., SHAH, S.D. & CRYER,P.E.
(1980) Epinephrine plasma metabolic clearance rates and
physiological thresholds for metabolic and haemodynamic
actions in man. Journal of Clinical Investigation, 66,94-101.
I131 CRYER,P.E., RIZZA,R.A., HAYMOND,
M.W. & GERICH,J.E.
(1979) Epinephrine and norepinephrine are cleared through
/3-adrenergic but not 6-adrenergic mechanisms in
man (Abstract). Clinical Research, 2 7 , 7 0 0 ~ .
I141 BROWN,M.J. (1981) Sensitivity and accuracy of phaeochromocytoma diagnosis. Lancet, i, 493-494.
I151 HAMILTON,W.F. (1962) Circulation; Handbook of
Physiology, chapter 39-44. Ed. Dow, P. American Physiological Society, Washington, D.C.
1161 GOODALL,McC. (1951) Studies of adrenaline and noradrenaline in mammalian hearts and suprarenals. Acta
Physiologica Scandinavica, 24 (Suppl. 85), 1-5 1.
171 SEDVALL,G. (1963) Noradrenaline storage in skeletal muscle.
Acta Physiologica Scandinavica, 60,39-50.
I181 YAMAGUCHI,
N., DE CHAMPLAM, J. & NADEAU,R.A. (1975)
Correlation between the response of the heart to sympathetic
stimulation and the release of endogenous catecholamines into
the coronary sinus of the dog. Circulation Research, 36,
662-668.
1191 COUSINEAU,D., FERGUSON,R.J., DE CHAMPLAIN,J.,
GAUTHIER,
P., COTE, P. & BOURASSA,M. (1977) Catecholamines in coronary sinus during exercise in man before and
after training. Journal of Applied Physiology, 43,801-806.
1201 D E QUATTRO,C. & CHAN,S. (1972) Raised catecholamines in
some patients with primary hypertension. Lancet, i, 806-809.
1211 CHRISTENSEN,
N.J. & BRANDSBORG,
0. (1973) The relationship between plasma catecholamine concentration and pulserate during exercise and standing. European Journal of
Clinical Investigation, 3,299-306.