Clinical Science and Molecular Medicine (1978). 55,23 1-234
A comparison of methods for measurement of rectal
potential difference in man: effects of rectal infusion of
amiloride
F. M O H A M E D , P. M c S O R L E Y
AND
D. J . W A R R E N
Department of Renal Medicine, University of Southampton, St Mary’s Hospital, Portsmouth, U.K.
(Received 10 November 1977; accepted 6 April 1978)
Summary
1. Two methods for measurement of rectal
potential difference in man were compared. A salinefilled catheter technique gave more reproducible
results and was better tolerated by patients than the
solid probe technique.
2. Infusion of amiloride at a concentration of 4
x
mol/l in saline produced a variable fall of
potential difference in six normal subjects. Doseresponse curves in two subjects showed that
complete inhibition of rectal potential difference occurred at 4 x lop5 mol/l and 4 x
moll1
respectively.
3. This finding provides additional evidence that
rectal potential difference in man results from
electrogenic ion transport across the mucosal
epithelial cell.
Key words: amiloride, electrogenic ion transport,
potentiometry .
Introduction
The colonic mucosa is electrically polarized, the
luminal side being negative with respect to the
serosal side (Cooperstein & Brockman, 1959). The
potential difference (PD) increases when sodium
transport is stimulated (Edmonds, 1967; Edmonds
& Marriott, 1967). In early studies in man
(Edmonds & Godfrey, 1970) subcutaneous electrodes were used and sigmoidoscopy was necessary
to insert the rectal probe. Endoscopy was later
Correspondence: Dr David J. Warren, Department of Renal
Medicine, University of Southampton, St Mary’s Hospital,
Portsmouth, Hants, U.K.
eliminated (Edmonds & Richards, 1970). and
saline infused into the rectum before a blunt-probe
electrode was advanced blindly.
Amiloride inhibits electrogenic ion transport in
animal models (Cuthbert, 1973), and our study in
man was designed to throw light on the mechanism of generation of rectal PD by observing the
effects of rectal administration of amiloride. In
addition the original endoscopic method for inserting a dry rectal probe has been compared with a
technique which eliminates endoscopy and depends
upon the introduction of a saline-filled catheter into
the rectum to provide electrical contact.
Methods
Informed consent was obtained from male members of the medical and laboratory staff, all of
whom had normal blood pressure and ate a normal
diet. Their ages ranged from 28 to 40 years. Rectal
PD measurements were made at the same time of
day on different days.
The endoscopic rectal P D method (Archampong
& Edmonds, 1972) required a probe electrode
which was built into the tip of a Perspex catheter
filled with agar-saline to provide electrical contact.
The reference (skin) electrode was a silver/silver
chloride junction fitted into a Perspex disc containing agar-saline. The probe was introduced through
a proctoscope and PD recorded on a highimpedance millivoltmeter.
The saline-filled catheter could be inserted into
the rectum without prior proctoscopy. The rectal
electrode (Fig. 1) consisted of a silver rod housed in
polyvinyl chloride (PVC) tubing through which
23 1
F. Mohamed. P. McSorley and D. J. Warren
asymmetry between them was no more than k 3
mV. Rectal PD was measured in six subjects, both
To millivoltmeter
the endoscopic technique with its dry electrode and
the saline-filled catheter technique being used alternately on the same occasion.
Nylon
Y-piece
To determine the effect of amiloride on rectal PD
Insulated
in six normotensive subjects a baseline was first
copper
wire
established by monitoring rectal PD for 20 rnin at
PVC tubing
30 s intervals whilst constantly infusing saline into
the rectum at 0.85 ml/min. On a subsequent
occasion rectal P D was measured during infusion
of amiloride. There was no response to a dose of 1
x
mol/l (Cuthbert, 1973) but a dose of 4 x
AgIAgCI
mol/l in isotonic saline was used to produce a
electrode
significant effect, being infused at 0.85 ml/min for
about 12 rnin after 8 rnin of control observations
and PD being measured at 30 s intervals. Rectal
FIG. 1. Rectal catheter assembly. The soldered joint between
PD was also measured in two normal subjects
the copper wire and the electrode is insulated by a short length
during
infusions of amiloride at concentrations
of PVC tubing sealed at both ends with epoxy resin (Araldite).
ranging from 4 x
to 4 x
mol/l, different
doses being infused at intervals of not less than 5
isotonic sodium chloride solution (154 mmol/l;
days.
saline) was infused continuously at 0.85 ml/min.
Results are expressed as mean & SD.
The rectal catheter was of narrower bore (internal
diameter 4 mm), fitted snugly into the end of the
PVC tubing which housed the electrode, and could
Results
be inserted 10 cm into the rectum. In order to avoid
junctional potentials, caused by saline coming into
Comparison of rectal potential difference methods
contact with the soldere$ joints, the latter were
The saline catheter technique for measurement
housed in plastic tubing sealed with epoxy resin
of rectal PD gave higher values in six normal
(Araldite). The skin electrode was a square sheet of
subjects (54.9
15.4 mV) than the endoscopic
silver (15 mm x 15 mm) with a narrow handle
technique (35.9 f 12.2 mV). In six normal
which was soldered to copper wire leading to the
subjects the coefficient of variation for six to1 1
millivoltmeter, the soldered junction being insulated
successive measurements with the endoscopic
as for the rectal electrode. Both electrodes were
technique ranged from 2.5 to 11.9% and from 1.0
coated with a layer of silver chloride by electrolysis
to 3.4% in the same subjects when the saline-filled
in hydrochloric acid (0.1 mol/l) with a 1.5 V cell.
catheter technique was used. The saline catheter
Subjects were studied in the left lateral position,
was better tolerated by the patients than the
without prior rinsing of the rectum. The rectal
endoscopic technique because of the narrow bore
catheter was primed with isotonic saline ensuring
that there were no air bubbles in the infusion
of the catheter.
system, and inserted 10 cm into the rectum. The
saline flow rate was increased to 8.5 ml/min for a
Atniloride infusions
few seconds to eliminate any bubbles which may
have formed at the time of insertion of the
In all six subjects a preliminary study using the
saline-filled catheter confirmed the stability of
catheter into the rectum, and the flow rate was
rectal PD over 20 min. Variations in PD in the first
subsequently maintained at 0.85 ml/min. The
2-3 min resolved spontaneously and the coefficient
reference electrode was wrapped in paper tissue,
of variation during the 20 rnin of rectal PD
wetted with isotonic saline and taped over the site
measurement was less than 5% when these initial
of an intradermal injection of 0.2 ml of isotonic
variations were excluded. Infusion of 4 x
mol
saline in the skin of the buttock. In preliminary
of amiloride/l in saline reduced the rectal potential
experiments rectal PD readings were taken every
difference by 50-100% in all six normal subjects.
30 s for 3 min. At the beginning and end of each
The log normal dose-response curve in two
series of measurements the two electrodes were
subjects showed that this was a dose-related
placed together in isotonic saline to check that the
232
Isotonic
Rectal potential difference
233
log [Amiloridel (mol/l)
FIG. 2. (a) Log normal dose-response curves in two subjects showing the response of rectal potential difference to
rectal infusion of amiloride. (b)
~,Least-souares regression lines derived from the ooints on the dose-resoonse curves.
The regression equation for the four-point line is y = -88x .- 471, and for the three-point h e y = -43x - 192.
-
response. Complete inhibition of rectal PD occurred in subject D.W. when amiloride was infused at
a concentration of 4 x
mol/l, and in subject
F.M. at 4 x
mol/l. The points on the two log
dose-response curves (Fig. 2a) were fitted by a
linear least-squares regression line (Fig. 2b). The
pooled data for each regression line showed that
the lines were significantly different from zero
response, but were not parallel.
Discussion
Determination of rectal potential difference in man
We found the endoscopic technique to be less
reproducible than the saline-filled catheter technique. Similar variability of endoscopic technique has
been noted by Beevers, Morton, Tree & Young
(1975), who found that the coefficient of variation
for 10-20 repeated measurements on eight subjects
ranged from 7.7 to 23%. Factors contributing to
the poor precision of the endoscopic technique
included damage and contamination of the agar
with faeces, since the rectum was not washed out
before each measurement. To obtain reproducible
results with the catheter it was necessary to ensure
that it was free of bubbles, which could give rise
to false or unstable values. The higher PD obtained
with the catheter may reflect the wide spread of
infused saline throughout the rectum, whereas
electrical contact with the solid probe is confined to
one point at which the mucosa may be easily
damaged by the probe. The saline-filled catheter
technique was better tolerated by our subjects and
had the additional advantage that it could be used
for pharmacological experiments requiring the
infusion of fluids into the rectum.
Effect of arniloride on rectal potential difference
Amiloride hydrochloride was infused into the
rectum to show whether rectal potential difference
in man was generated by electrogenic ion transport.
It inhibits such transport in frog skin (Cuthbert,
1973), amiloride at 1 pmol/l producing 90%
inhibition of short-circuit current. In man, however,
we found it necessary to use a concentration of 4
pmol/l in order to obtain a marked effect. There
was a dose-dependent relationship between the
concentration of infused amiloride and the extent of
inhibition of rectal potential difference. The difference between slopes of the dose-response curves
could be due to a combination of factors including
variation in receptor sensitivity or the concentration of the amiloride at the biophase. The
concentration of infused amiloride may well not be
the concentration at the receptor site since the
infusate may be diluted by fluid already in the
rectum.
Active transport of sodium occurs from the
lumen of the rectum to the blood against a
concentration gradient (Edmonds, 197 1). This
appears to be largely responsible for the observed
transmucosal potential difference (Cooperstein &
Hogben, 1959; Archampong & Edmonds, 1972).
An increase in human rectal potential difference is
associated with the increased sodium absorption
which follows aldosterone and carbenoxolone
administration (Edmonds & Godfrey, 1970; Tomkins & Edmonds, 1975).
Amiloride has recently been shown to cause a
prompt fall in potential difference across human
colonic mucosa when applied to the mucosal
surface in vitro, and rectal infusion of amiloride
( I mol/l) in vivo reduced PD by 33-52% (Rask
Madsen & Hjelt, 1977). Our study confirms this
finding and also shows a dose-dependent relationship between rectal potential difference and
the concentration of amiloride infused into the
rectum. Reduction of short-circuit current and PD
across both human colonic mucosa in vitro, and
also isolated frog skin, in response to amiloride is
associated with reduction in sodium flux (Salako &
234
F. Mohamed, P. McSorley and D. J. Warren
Smith, 1970; Rask-Madsen & Hjelt, 1977). By
analogy we suggest that the effects of amiloride on
human rectal PD in vivo reflect a reduction in
sodium flux across the rectal mucosal cell.
Acknowledgment
We are grateful to Professor A. Polak for the
provision of laboratory facilities and criticism of
the manuscript, to Dr Brian Parsons for helpful discussion and Mr M. J. Emerson for statistical help.
References
ARCHAMPONG,
E.Q. & EDMONDS,
C.J. (1972) Effect of luminal
ions on the transepithelial electrical potential difference of
human rectum. Gut, 13,559-565.
BEEVERS,D.G., MORTON,J.J., TREE,M. & YOUNG,J. (1975)
Rectal potential difference in the diagnosis of aldosterone
excess. Gut, 16,36-41.
COOPERSTEIN, I.L. & BROCKMAN,
S.K. (1959) The electrical
potential difference generated by the large intestine; its
relation to electrolyte and water transfer. Journal of Clinical
Inuesrigatiori, 38,435-442.
I.L. & HOGBEN,C.A.M. (1959) Ionic transfer
COOPERSTEIN,
across the isolated frog large intestine. Journal of Clinical
Inuestigatiori, 38,46 1-473.
CUTHBERT,A.W. (1973) An upper limit to the number of
sodium channels in frog skin epithelium. Journal of Physiology (London),228,68 1-692.
EDMONDS,C.J. (1967) The gradient of electrical potential
difference and of sodium and potassium of the gut contents
along the caecum and colon of normal and sodium-depleted
rats. Journal of Physiology (Lotidon), 193,57 1-588.
EDMONDS,C.J. (1971) Absorption of sodium and water by
human rectum measured by a dialysis method. Gu!, 12,356362.
EDMONDS,
C.J. & GODFREY,R.C. (1970) Measurement of
electrical potentials of the human rectum and pelvic colon in
normal and aldosterone treated patients. Gut, 11,330-337.
EDMONDS,C.J. & MARRIOTT,J.C. (1967) The effect of
aldosterone and adrenalectomy on the electrical potential
difference of rat colon and on the transport of sodium, potassium. chloride and bicarbonate. Journal of Endocrinology,
39,5 17-53 I.
EDMONDS,
CJ. & RICHARDS,
P. (1970) Measurement of rectal
electrical potential difference as an instant screening test for
hyperaldosteronism. L a w e t , ii, 624-627.
RASK-MADSEN,
J. & HJELT,K. (1977) Effect of amiloride on
electrical activity and electrolyte transport in human colon.
Scatidinauian Journal of Gastroenrerology, 12, 1-6.
SALAKO,L.A. & SMITH,A.J. (1970) Effects of amiloride on
active sodium transport by the isolated frog skin: evidence
concerning site of action. British Journal of Pharmacology,
38,702-718.
TOMKINS, A.M. & EDMONDS,
C.J. (1975) Electrical potential
difference, sodium absorption and potassium secretion by the
human rectum during carbenoxolone therapy. Cur, 16, 277284.