Clinical Science (1996) 91, 23-28 (Printed in Great Britain)
23
Smooth musclederived nitric oxide is elevated in isolated
forearm veins in human alcoholic cirrhosis
Jeremy RYAN, Garry JENNINGS, Frank DUDLEY and ]aye CHIN-DUSTING
Department of Gastroenterology and Alfred and Baker Medical Unit, Alfred Hospital and Baker
Medical Research Institute, Melbourne, Victoria, Australia
(Received 12 January/l4 March 1996; accepted 19 March 1996)
1. Cirrhosis is often complicated by disturbances in
the systemic circulation. We have previously demonstrated decreased vascular responses to vasoconstrictors in forearm resistance arteries in subjects with
alcoholic cirrhosis. In the current study we investigate
the role of the potent endogenous vasodilator nitric
oxide in the peripheral circulation of these patients.
2. Ten patients with alcoholic cirrhosis (Pugh grade
A) and 10 age-matched control subjects were studied.
The effect of blockade of nitric oxide synthesis was
studied both in vivo in forearm resistance arteries
using forearm venous occlusion plethysmography and
in vitro in veins isolated from the forearm. The role
of endothelium-derived nitric oxide was studied in
vivo using the endothelium-dependent vasodilator
acetylcholine.
3. Mean arterial pressure and forearm basal flow in
vivo were similar in the two groups. The constrictor
response (percentage decrease in forearm blood flaw)
to noradrenaline (100 nglmin) was 26% smaller in
patients with cirrhosis (31.65f 2.64%) than in control
subjects (42.75 f 3.87%, P= 0.037). Constrictor responses to the nitric oxide synthase inhibitor
NC-monomethyl-L-arginine were not different in the
two groups. Dilator responses to acetylcholine were
significantly attenuated in cirrhotic patients compared
with control subjects.
4. To investigate the role of smooth muscle-derived
nitric oxide in vitro, all veins were stripped of their
endothelium. Responses to noradrenaline were significantly diminished in veins isolated from patients with
cirrhosis compared with control subjects. Incubation
with the nitric oxide synthase inhibitor R"nitro-L-arginine had no effect on responses to noradrenaline in veins from control subjects but significantly enhanced the maximal response to noradrenaline by 23.95% (range 3.77-loo%, P-0,043) in veins
from patients with cirrhosis.
5. Responses to noradrenaline were attenuated in vivo
in forearm resistance arteries in patients with alcoholic cirrhosis. This impairment was also apparent in
forearm isolated veins, stripped of the endothelium.
Our data exclude a major role for endothelium-
derived nitric oxide but highlight a possible role far
smooth muscle-derived nitric oxide.
INTRODUCTION
Vascular smooth muscle tone is maintained by a
complex interplay of both endotheliram-dependent
and independent factors [l], These factor8 have the
potential to either constrict or relax vascular
smooth muscle, and the resultant resistance to flow
is dependent on the overall balanm af theso aompet.
ing stimuli, We have previously dmnanstrated B
decreased vascular responsiveness to bsth nsradrenaline and angiotensin I1 [2ll de6pits an increase in
sympathetic nervous system activity €33 and aotiva:
tion of the ranin-angistenein ~ y ~ t e min, patients
with alcoholic cirrhosis, A psssible mplanatien f ~ r
this impaired respansiveneas i~
that synthods QF tho
endogenous vasodilator nitria elride 18 induoed la
response to increased plasma endotoxin levels [43.
Support for this proposal is the demonstration that
patients with cirrhosis have significantly higher
serum nitrite and nitrate levels and that the levels of
these metabolites of nitric oxide correlate well with
plasma endotoxin levels [S]. It has also been shown
that blockade of nitric oxide synthesis in rats with
carbon tetrachloride-induced cirrhosis reverses the
vascular disturbances otherwise observed [6-81.
In the current study, we examine the effects of
nitric oxide synthesis blockade in vivo in forearm
resistance arteries, as well as in uitro in veins
isolated from cirrhotic patients. In addition, the role
of endothelium-derived nitric oxide was studied in
uiuo using the endothelium-dependent vasodilator
acetylcholine.
MATERIALS AND METHODS
Subjects
Forearm venous occlusian plethysmogrsphy. The cirrhotic group consisted of 10 males with alcoholic
liver disease (mean age $7,0k3.19 years) recruited
from the Liver Clinic at the Alfred Hospital. Half
had histological confirmation of cirrhosis; the other
Key wordi: alcoholic cirrhosis, nitric oxide, resistance arteries, veins.
Abbrevlatlonfi: t-NMMA, NG-rnonomethyl-c-argininei MANQVA, multivariate analysis of variance: MAP, mean merial presrure; NOLA, nirrc+c-arginine,
Correrpondencel Dr J,Chin-Dusting, Alfred and Baker Medical Unit, Baker Madical Rerearch lnditufe, Melbourne, Victoria 3181, Australia,
24
J. Ryan et at.
half were diagnosed clinically and on evidence of
portal hypertension (presence of oesophageal varices). All were classified according to the criteria of
Pugh et al. [9] as Pugh grade A. At the time of
study none had the clinical features of decompensation (ascites, oedema or encephalopathy) or of peripheral neuropathy. Diuretics and all other vasoactive medications were stopped at least 5 days before
the day of study. All subjects were on a normal diet,
containing approximately 100 mmol of sodium per
day. Mean plasma cholesterol and triacylglycerol
levels were 4.44k0.31 and 1.10~0.15mmol/l
respectively.
The control group consisted of 10 healthy, agematched (mean age 55.6k3.53 years) men with no
evidence of liver disease on history, physical examination or biochemical tests. All were negative for
the hepatitis B surface antigen and hepatitis C
antibody (by ELISA). Mean plasma cholesterol and
triacylglycerol
levels
were
5.23 k0.27
and
1.51 k 0.28 mmol/l respectively and not significantly
different from those of cirrhotic patients. Control
subjects were recruited by advertisement and an
out-of-pocket expenses stipend (75 Australian
dollars) was offered to each control subject.
Forearm vein biopsy. All subjects were invited to
return for a vein biopsy. Six cirrhotic patients and
four control subjects agreed to this procedure.
The project was approved by the Alfred Group of
Hospitals Ethics Committee. All subjects gave their
written, informed consent and were aware of their
right to withdraw from the study at any time of
their choosing.
Procedures
Forearm venous occlusion plethysmography. Studies were performed in a quiet room, maintained at
a temperature of 22°C. All subjects were asked to
refrain from consuming caffeine-containing beverages for 12 h before the study. All vasoactive medication was stopped at least 5 days before the day of
study. Subjects were asked to abstain from alcohol
for 5 days before attendance.
Forearm vascular responses to the vasoactive
substances used were measured on the left arm. The
left brachial artery was cannulated (3.0-F, 5-cm
catheter; Cook, Australia) under local anaesthesia
(1% lignocaine; Astra, NSW, Australia) and full
aseptic conditions. Intra-arterial blood pressure was
recorded with a disposable physiological pressure
transducer (model CMS-327; Biosensors International, Singapore, linked with Spacelabs, Washington,
DC, U.S.A.). Forearm blood flow was measured by
venous occlusion plethysmography with a sealed,
alloy-filled (gallium and indium), double-strand
strain gauge (Medasonic, Mountain View, CA,
U.S.A.) and recorded for 10 out of every 20s.
Venous occlusion pressure was 4G50mmHg at the
proximal (elbow) end and cuff occlusion pressure at
the distal (wrist) end approximated 200 mmHg.
Forearm vascular resistance ( R ) was calculated by
the formula:
R =mean arterial pressure (MAP; mmHg)/
forearm blood flow (ml min-' lOOml-').
Protocol. Responses to local, sequential infusions
of the endothelium-dependent vasodilator acetylcholine (9.25, 18.5 and 37 pg/min), noradrenaline
(100ng/min) and the nitric oxide synthase inhibitor
NG-monomethyl-L-arginine ( L-NMMA; 1, 2 and
4 mmol/min) were obtained. After an initial equilibration period of 60s, the average of three flow
measurements before each drug infusion was
obtained and used as a measure of basal blood flow.
Each drug concentration was infused at 2 ml/min
over a minimum of 2min (for acetylcholine and
noradrenaline) or 5min (for L-NMMA) or until the
response over three flow measurements reached a
plateau. The average of these three flow measurements was obtained as a measure of drug-induced
flow. Rest periods of 5-l0min between concentrations and 15min between drugs were allowed.
When a plateaued effect of L-NMMA (4 pmol/min)
had been achieved, the response of a simultaneous
infusion with acetylcholine (37 pg/min; for a minimum of 2min or until a plateaued response over
three flow measurements) was obtained. This was
followed by washout of acetylcholine (5 min), after
which responses to a simultaneous infusion of noradrenaline (lOOng/min; for a minimum of 2min or
until a plateaued response over three flow measurements) together with L-NMMA (4 pmol/min) were
obtained. Thus, L-NMMA (4 pmol/min) was infused
over a period of 18-20min in total. These drug
concentrations had no effect on either systemic
blood pressure or heart rate measured by an electrocardiogram lead I1 (Spacelabs).
Vein biopsy. Forearm vein biopsies were performed on a separate day. Biopsies were performed
under local anaesthesia (lignocaine, 1% subcutaneously) with strict aseptic conditions. An incision
of approximately 1.5cm in length was made at the
radial border of the wrist. When a vein was found,
two ties were made with a 1-cm interval. The
segment of vein between the two ties was dissected
and removed. The isolated veins were immediately
placed in ice-cold Krebs-Henseleit solution (composition, mmol/l: NaCl 119, KCI 4.7, KH,PO, 1.18,
MgSO, 1.17, NaHCO, 25, CaCl, 2.5, EDTA 0.026,
glucose 11) and transported to the laboratory,
where the isolated vein was carefully cleared of
connective tissue and dissected into a 3-mm-wide
ring. Denudation was performed by carefully rubbing the lumen around a 2-mm-diameter stick.
Rings were then mounted onto two L-shaped parallel wires (355pm diameter) in a 25-ml jacketed
organ bath and bathed in Krebs solution maintained at 37°C and bubbled with carbogen (95%
oxygen, 5% carbon dioxide). After an unstretched
Nitric oxide and alcoholic cirrhosis
period of 30 min, each vessel was passively stretched
to an internal circumference equal to 0.9 x L,,
standard pressure, where L,, denotes the internal
circumference at the level of passive stretch equivalent to a transmural pressure of 20mmHg [lo]. This
was calculated using the length-tension relationship
obtained for each vessel and was applied to ensure
that veins of different internal diameter were
stretched to a similar point on their length-tension
curve, thus enabling normalization of the results
obtained.
After a further 30min, the viability of each vessel
was tested by addition of noradrenaline (30 nmol/l).
When a plateaued constriction was obtained, successful denudation of the endothelium layer was
confirmed by addition of the endotheliumdependent vasodilator, acetylcholine (1 pmol/l).
Fifteen minutes after washout of noradrenaline
and acetylcholine, a full concentration-dependent
curve to noradrenaline was obtained. This was then
repeated after 30min incubation and in the continued presence of the nitric oxide synthase inhibitor,
nitro-L-arginine (NOLA; 10pmol).
Analysis. Unless otherwise stated, results are given
as means fSEM and analysed using Student's t-test
(paired where appropriate). P < 0.05 was taken as
the criterion for statistical significance.
Forearm venous occlusion plethysmography. We
have previously found a high within-subject coefficient of variation (range 0.02-0.47; mean 0.19 f0.02)
[l I]. Hence, the response to each concentration of
the agonists was normalized to the basal forearm
blood flow obtained immediately before the administration of each dose. The effects on forearm
blood flow of acetylcholine and L-NMMA were
compared in the two groups by multivariate analysis of variance (MANOVA; SPSS/PC Statistical
data analysis; SPSS, Chicago, IL, U.S.A.).
Vein biopsy. Contractile responses were measured
as force (g)normalized to the internal diameter of
each ring. Individual concentration-response curves
were fitted to a logistic equation of the form
E = MAP/(AP x K P ) , where E is the response, M is
the maximum response, A is the concentration
eliciting E, K is the concentration eliciting 50% of
the maximum response (i.e. EC,,) and P is the slope
parameter [12]. The EC5, value provided a measure
of sensitivity.
RESULTS
In vivo forearm vascular responses
Mean forearm basal flow was not different in the
two groups (cirrhotic versus control subjects:
2.48f0.83mlmin-' 100ml-' versus 2.12k0.35ml
min-' lOOml-'; P>O.O5). Mean arterial pressure
(cirrhotic versus control subjects: 91.52 f2.63 mmHg
versus 84.08 f2.90 mmHg; P >0.05) and (calculated)
forearm vascular resistance units (cirrhotic versus
control subjects: 78.39 f20.74 versus 54.15 f 10.78
25
Acetylcholine (pglmin)
20
30
50
40
0
8-I0 20
~~
\\
B
S -30-
.-1
n
e
-b
-40-
z -503
E
L
e
P = 0.049
-60-70
-
-80
_I
Fig. I. Acetylcholine responses in forearm resistance arteries. Responses (percentage decrease in forearm vascular resistance) to acetylcholine
(9.25, 18.5 and 37pglmin) were obtained in forearm resistance arteries
using venous occlusion plethysmography. Analysis by multivariate analysis of
variance (MANOVA), followed by Student's t-test confirmed that these
n = 10)
responses were significantly attenuated in the cirrhotic patients
compared with agematched control subjects (0,n = 10). Error bars
represent standard error of the difference.
(m,
resistance units, P > 0.05) were similarly not different
in the two groups.
Forearm vascular resistance decreased dosedependently with acetylcholine in both groups. Responses to acetylcholine were significantly attenuated
in forearm resistance arteries of cirrhotic patients
compared with the control group (MANOVA
P=0.049; Fig. 1).
Noradrenaline ( 100ng/min) decreased forearm
blood flow, i.e. increased forearm vascular resistance, in both groups. However, the response to
noradrenaline in cirrhotic forearm vessels was significantly smaller than the response obtained in blood
vessels from the control group (decrease in forearm
blood flow, cirrhotic versus control subjects:
31.64+2.64% versus 42.75+3.87%, P=O.O37).
L-NMMA dose-dependently increased forearm
vascular resistance in both groups (Fig. 2). There
was no significant difference between the two groups
(MANOVA, P=O.94).
In control subjects, L-NMMA (4 pmol/min) significantly attenuated the response to acetylcholine
(37 pmol/min; P = 0.01) and significantly enhanced
the response to noradrenaline (100 ng/min; P =0.03)
(Fig. 3). Although the same trends were seen for
responses to both these agonists in the cirrhotic
subjects, L-NMMA did not significantly alter these
responses (Fig. 3).
In vitro venous responses
Vein lumen size (diameter at 20mmHg) was not
significantly different in the two groups (Table 1). A
1.
Ryan et al.
1
0
I
1
3
4
5
t-NMMA (pmol/min)
Fig. 2. L-NMMA responses in forearm resistance arteries. Responses
(percentage increase in forearm vascular resistance) to the nitric oxide
synthare inhibitor N6-monomethyl-t-arginine (t-NMMA I , 2 and 4pmol/
min) were obtained in forearm resistance arteries using venous occlusion
plethysmography. These responses were not different in the cirrhotic group
n= 10) compared with control subjects (0,n = 10). Analysis was by
MANOVA. Error bars represent standard error of the difference.
(m,
lack of response to acetylcholine (1 ,umol/l) on vessels contracted with noradrenaline confirmed that
each ring had been successfully stripped of the
endothelium (results not shown). Maximal responses
to noradrenaline were significantly attenuated in
vein rings from cirrhotic patients compared with
those from control subjects (Fig. 4). Potency (EC,,)
to noradrenaline was not different in veins from
cirrhotic patients compared with controls (Table 1).
NOLA (lO,umol/l) had no effect on either the
potency (EC,,) or the maximal response (FmaX)
to
noradrenaline in veins from control subjects (Table
1). Although potency to noradrenaline was similarly
unaltered by NOLA (Table 1) in veins from cirrhotic patients, maximal responses to noradrenaline
increased by an average of 23.95% (range 3.77100%; Table 1). Veins from control subjects
increased by an average of 1.56% (range &2.95%).
DISCUSSION
The primary finding of the current study was that
impaired responses to noradrenaline in forearm
resistance arteries from cirrhotic patients in vivo
([2]; present study) were also apparent in
endothelium-denuded veins isolated from these
patients. The blunted response to noradrenaline in
the isolated veins could be partly reversed by incubation with the nitric oxide synthase inhibitor
NOLA. No similar alteration was apparent in veins
from the control group. These data suggest that the
endothelium, or substances from the endothelium,
do not contribute to the blunted responses
observed. This excludes a major role for
endothelium-derived nitric oxide and prostanoids
such as prostacyclin, but highlights a possible role
for smooth muscle-derived nitric oxide.
Our data are in direct contrast to the findings of
Castro et al. [6], who showed that endothelium
denudation reversed the blunted pressor response in
aortic rings from rats with carbon tetrachlorideinduced cirrhosis. As the nitric oxide synthase inhibitor NOLA produced a similar reversal in their
study, it was suggested that endothelial nitric oxide
contributes to the depressed response to constrictor
agonists However, in the patients with alcoholic
cirrhosis in the current study, the augmented responses to noradrenaline after NOLA incubation
occurred in the absence of an endothelium and,
although small, were consistent, occurring in each of
the six cirrhotic veins studied. We thus suggest that
smooth muscle-derived, and not endotheliumderived, nitric oxide contributes to the impaired
pressor response in cirrhosis.
In the current study, basal forearm blood flow in
cirrhotic subjects was not different to that in agematched control subjects. This finding was similar
to the results of our previous study [2], but contrasts with the recent findings of Calver et al. [13].
This may be indicative of the well-compensated
nature of the cirrhotic patients in the current study.
Again mirroring results from our previous study, in
which a three-point (25, 50 and 100ng/min) doseresponse curve to noradrenaline was constructed,
the single-dose ( 100ng/min) response to noradrenaline was significantly blunted in the forearm resistance arteries of cirrhotic patients. A depressed
pressor response is well documented in animal
models of cirrhosis [6, 81 as well as in humans [14,
151. On the other hand, the constrictor responses in
viuo of forearm resistance vessels to the nitric oxide
inhibitor L-NMMA were not different in cirrhotic
patients compared with control subjects. This finding argues against widespread arteriovenous shunts
in cirrhotic vessels that may be unresponsive to
pressor stimuli as this would be expected also to
reduce responses to L-NMMA. The inhibition of
nitric oxide activity by L-NMMA is unselective,
blocking both endothelium and smooth muscle
nitric oxide synthase. In forearm resistance arteries
of control subjects, L-NMMA significantly inhibited
responses to acetylcholine, demonstrating some
antagonism at the endothelium level, and significantly augmented responses to noradrenaline. The
latter indicates that basal release of nitric oxide
occurs in this vascular bed and modulates responses
to vasoconstrictors. As the response to L-NMMA
was not different in cirrhotic patients and control
subjects, it follows that excess nitric oxide production was not responsible for the blunted response to
noradrenaline. An unexpected but interesting finding
was the diminished response to the endotheliumdependent relaxing agonist acetylcholine in forearm
vessels of cirrhotic patients. Without studying the
effects of other endothelium-dependent and -independent vasodilators, however, it is difficult to know
Nitric oxide and alcoholic cirrhosis
27
NS
*P <0.05
2oo
T
1
Acetylcholine
L-NMMA
Noradrenaline
-75
,
-I
- . .. .. .. .
NS
Coitrol
-la0 J
L-NMMA
*P <0.05
*P <0.05
Fig. 3. Effect of L-NMMA on responses to acetylcholine and noradrenaline in forearm resistance arteries. Both dilatory responses to acetylcholine (37pglmin)
and constrictor responses t o noradrenaline (100ng/min) were significantly diminished in forearm resistance arteries of cirrhotic patients compared with those of control
subjects. Responses to acetylcholine (37pg/min) were significantly blunted by co-infusion with the nitric oxide synthase inhibitor t-NMMA (4pmol/min), whereas
responses to noradrenaline (100nglmin) were significantly enhanced in forearm resistance arteries of control subjects (n = lo). Although similar trends were observed for
cirrhotic patients, these values did not achieve statistical significance.
Table 1. Effect of NOLA on potency (negative log EC,,) and sensitivity (maximal response; F,,,) to noradrenaline in
endotheliumdenuded veins isolated from control ( n 4 ) and cirrhotic ( n d ) subjects. *P <0.05. Paired Student’s t-test
compared with value before NOLA.
_____
~
Noradrenaline before IOpmol/l NOLA
Noradrenaline after IOpmol/l NOLA
EC,,
Fm,
F,,/diameter
(neg. log m 4 )
(9)
k/mm)
EC,,
beg. 1% mol/l)
Fmax
Group
Vein
diameter
(mm)
(9)
F,,/diameter
Umm)
Control
Cirrhotic
3.81 kO.55
3.04 f 0.40
6.96k0.22
7.08 f 0. I4
8.01 f2. I2
3.57f 1.38
2.09 k0.32
1.35f0.54
6.94 k0.20
7.02f0.17
8. I6 y 2.22
3.99 & I .49*
2. I I & 0.47
I.49 & O M *
if this is indicative of altered endothelial nitric oxide
production in this patient group. Another explanation may be that a negative feedback regulation of
endothelial nitric oxide in the face of elevated nitric
oxide levels (due to induced nitric oxide from
smooth muscle cells) occurred. This has previously
been demonstrated in bovine aortic endothelial cells
C161.
In conclusion, we confirm that patients with wellcompensated alcoholic cirrhosis have an abnormality i.n nitric oxide production. It is clear, however,
that this abnormality is somewhat more complex
than originally proposed by Vallance and Moncada
and that changes in local nitric oxide release are not
the only factor contributing to the hyporesponsiveness to vasoconstrictors observed in these subjects.
ACK N0W LEDGMENTS
We thank Dr Krishnankutty Sudhir for expert
advice in the conduct of the in uivo experiments, Dr
Sue Cromie for recruitment of the cirrhotic patients
28
j. Ryan
2.5
1
/'
*P=0.021
MANOVA
-l-4Lnl
-10
al.
2. Ryan J, Sudhir K, Jennings G, Esler M, Dudley F. Impaired reactivity of the
/
0.0
et
-9
-8
-7
-6
-5
-4
Log [noradrenaline] (mol/l)
Fig. 4. Noradrenaline responses i n isolated veins. Full concentration-response (normalized to vein diameter) curves to noradrenaline were
obtained in veins isolated from the forearms of cirrhotic (m, n=6) and
control (0,n=4) subjects. All veins were stripped of their endotheliurn.
Analysis by MANOVA confirmed that responses to noradrenaline were
significantly blunted in veins from cirrhotic patients compared with agematched control subjects. Error bars represent standard error of the
difference.
and Mrs Pam Arnold for technical expertise in the
conduct of the in vitro experiments. This work was
supported by a project grant from the Alfred Hospital Research Trusts and an institute grant awarded
to the Baker Medical Research institute from the
National Health and Medical Research Council of
Australia.
REFERENCES
I. Moncada S, Palmer, RMJ. Higgs EA. Nitric oxide: physiology, pathophysiology
and pharmacology. Pharmacol Rev 1991; 4 3 10942.
peripheral vasculature to pressor agents in alcoholic cirrhosis.
Gastroenterology 1993; 105 1167-72.
3. Esler M, Jennings G, Korner P, et al. The assessment of sympathetic nervous
system activity in man from measurements of norepinephrine turnover.
Hypertension 1988; II: 3-20.
4. Vallance P, Moncada S. Hyperdynamic circulation in cirrhosis: a role for nitric
oxide! Lancet 1991; 337: 776-8.
5. Guarner C, Soriano G, Tomas A, e t al. Increased serum nitrite and nitrate
levels in patients with cirrhosis: relationship to endotoxemia. Hepatology 1993;
18: 1139-43.
6. Castro A, Jimenez W, Claria J, et al. Impaired responsiveness to angiotensin II
in experimental cirrhosis: role of nitric oxide. Hepatology 1993; 18: 367-72.
7. Pizcueta P, Pique JM, Fernandez M, et al. Modulation of the hyperdynamic
circulation of cirrhotic rats by nitric oxide inhibition. Gastroenterology 1992;
103 1909-15.
8. Sieber CC, Lopez-Talavera JC, Groszmann RJ. Role of nitric oxide in the in
vitro splanchnic vascular hyporeactivity in ascitic cirrhotic rats.
Gastroenterology 1993: 104: 1750-4.
9. Pugh RNH, Murray-Lyon M, Dawson JL, Pietroni, MC, Williams R. Transection
of the oesophagus for bleeding oesophageal varices Br J Surg 1973; 60:
6468.
10. Sudhir K, Angus ]A, Esler MD, Jennings GL, Lambert GW, Korner, PI. Altered
venous responses to vasoconstrictor agonists and nerve stimulation in human
primary hypertension. J Hypertens 1990 8: I 119-28.
I I. Chin J, Gust A, Nestel P, Dart, A. Marine oils dosedependently inhibit
vasoconstriction of forearm resistance vessels in humans. Hypertension 1993;
21: 224.
12. Chin JPF, Kaye DM, Hurlston RM, Angus JA, Jennings GL, Dart AM. Effects of
marine oil supplementation on reactivity of human buttock subcutaneous
arteries and forearm veins in vitro. 8r J Pharmacol 1994 1 1 2 56670.
13. Calver A, Harris A, Maxwell ID, Vallance P. Effect of local inhibition of nitric
oxide synthesis on forearm blood flow and dorsal hand vein size in patients
with alcoholic cirrhosis. Clin Sci 1994; 86: 203-8.
14. Lunzer MR, Newman SP, Bernard AG, Manghani KK, Sherlock SP, Ginsburg J.
Impaired cardiovascular responsiveness in liver disease. Lancet 1975; II:
382-5.
15. MacGilchrist AJ, Sumner D, Reid JL. Impaired pressor activity in cirrhosis:
evidence for a peripheral vascular defect. Hepatology 1991; 13; 689-94.
16. Buga GM, Griscavage JM, Rogers NE, lgnarro LJ. Negative feedback regulation
of endothelial cell function by nitric oxide. Circ Res 1993;7 3 808-12.