AMERICAN
JOURNAL
OF PHYSIOLOGY
Vol. 229, No. 5, November 1975.
Interstitial M.
Printed
in U.S.A.
fluid pressure in canine gastric
ALTAMIRANO,
Departamento
Caracas,
de Ciencias
M. REQUENA, AND TATIANA Fisiolbgicas, Escuela de A4edicina J. M.
interstitial
fluid
pressure;
gastric
muscle
activity
1963, GUYTON (7) introduced a new procedure to measure the interstitial fluid pressure (IFP). Small plastic capsules pierced by holes were implanted in a tissue a few weeks before the pressure of the fluid inside the capsule (ICP) was measured. With this method, the IFP of normal subcutaneous tissue and muscle appeared negative by about -7 mmHg. Later, the use of cotton strands to connect the interstitial fluid with the measuring device, i.e., the wick method, confirmed that at least in skin and muscle the IFP is usually negative (12, 15-18). Estimates of that pressure in the mucosa of hollow organs like the stomach, where muscle activity might modify the IFP, have not been published. In the stomach, furthermore, hydraulic gradients across the epithelium seem to cause some type of secretion ( 1, 2), which adds interest to the knowledge of the local IFP. To study this problem, we implanted in the submucosa of the corpus modified versions -of Guyton’s capsules (7). The intracapsular pressure (ICP) was taken as an accurate estimated of the IFP (9). In this report we will describe a number of tests carried out to evaluate the performance of the capsules used by us. The mean value of the ICP was significantly larger than that found by Guyton (9), but experimental factors might explain at least part of the difl’erence. For instance, the activity of the gastric muscles substantially modifies the ICP. It was also found that the
IN
Central
de Venezuela,
Venezuela
ALTAMIRANO, M., M. REQUENA, AND TATIANA C. PEREZ. Interstitialjluidpressure in canine gastric mucosa. Am. J. Physiol. 229(5) : 1414-l 420. 1975.~-Guyton’s capsules were implanted in the submucosa of the gastric corpus of dogs. The pressure of the fluid inside the capsule (ICP) was measured between 12 and 42 days later after mounting the piece of the corpus in a Plexiglas chamber. The capsule was always filled with saline. In two out of three experiments, the ICP did not change significantly when the saline was replaced by isotonic glucose or blood plasma. Changes of pressure exerted on the surface of the stomach were accurately monitored by the ICP recordings. Changes of circulation produced by compression of the artery or the vein connected to the piece of mucosa, or by intra-arterial injections of epinephrine, norepinephrine, isoproterenol, or hypertonic mannitol, modified the ICP as predicted by Starling’s law of capillary filtration. Spontaneous activity of the gastric muscles or the activity, which followed intra-arterial injections of acetylcholine, prostigmine, or histamine, changed the ICP significantly. Intra-arterial atropine usually decreased the ICP by 3-5 mmHg. The mean value of the ICP in 49 animals was 0.53 zt 0.34 mmHg (SEM); it was negative in 43y0 of the experiments.
gastric
C. P&REZ Vargas, Uniuersidad
mucosa
capsules monitored accurately stomach, without interference gastric con tents and/or gastric
the pressure inside the with the movements of the walls.
METHODS
Mongrel dogs of either sex weighing between 12 and 25 kg were used. A small Guyton capsule (9) was implanted under Pentothal anesthesia in the gastric corpus close to the great curvature. A number of different capsules were used: the most satisfactory model is illustrated in Fig. 1. A plastic stem of about 2 mm diam was introduced to the very end of the capsule through the tube shown in the picture. The tube itself measured about 4-6 cm in length. The stem was pulled out for the final experiment, assuring the presence of a cavity inside the capsule notwithstanding the complete healing of the operatory wound. Eventually, the capsule was connected to a small displacement pressure transducer (Micron Instruments, Inc.) through plastic tubing. The capsules were placed in the submucosa through a short incision of the muscle layer that was subsequently sutured. Within 3 wk the capsule became enveloped by a thin film of connective tissue, usually of less than 0.5 111111 of thickness. The upper part of Fig. 1 illustrates a section of fixed mucosa after the capsule was pulled out. The small stumps of tissue that appear inside the cavity penetrated into the capsule through the wall perforations and were connected to a layer of newly formed tissue applied to the capsule’s walls. Histological examination of 26 mucosae disclosed that the epithelium overlying the capsule was normal in most cases. In a few animals, it was somewhat thinner than the neighboring epithelium. In many animals the capsule perforated the epithelium and was lost or remained in the gastric cavity attached to the wall by the plastic tubing. The dogs were anesthetized with pentobarbital 12-42 days after the implantation of the capsules. A flap of the fundus was mounted in a Plexiglas chamber as described in a previous paper (2). The chamber used and illustrated in Fig. 2 permitted us to perform a variety of measurements. When completely filled with isotonic solutions, as shown in the figure, it worked as a plethysmograph: changes in volume of the mucosa resulted in movement of fluid into or from the tube which hung from the tension transducer. The corresponding variations in weight were recorded using a Grass polygraph. Changes of 0.1 g were accurately measured (see Fig. 10). In other experiments, the lateral tube of the chamber was connected to a pressure transducer,
1414
Downloaded from www.physiology.org/journal/ajplegacy at Glasgow Univ Lib (130.209.006.061) on February 13, 2019.
INTERSTITIAL
FLUID
PRESSURE
Il\r CANINE
GASTRIC
1415
MUCOSA
carefully aspirated at the end of each period into tared tubes and weighed again. The gastric contents were analyzed for Cl- (Chloridometer), Na+ and K+ (flame photometer), and H+ by titration of an aliquot to pH 7. Samples of plasma were also taken for Na+, K+, and Cl- determinations. The area of stretched mucosa was measured in each experiment as described previously (2). RESULTS
FIG. 1. A: section of gastric wall fixed with formaldehyde after pulling away implanted capsule. White arrows show thickness of mucosa over capsule. Layer of fibrous tissue around capsule measures about 0.5 mm in thickness. B: photograph of a capsule.
so that small changes of pressure inside the chamber could be recorded (see Figs. 3 and 4). Most of the experiments were performed without the plastic triangular plates that support the mucosa (Fig. 2) and with the chamber empty or containing fluid just sufficient to cover the mucosa. Systemic pressure was measured in a small artery of the hindlegs. In some experiments the gastric arterial pressure was modified by means of an extracorporeal circulatory system. One femoral artery was connected by a plastic tube to the artery which supplied the mucosal flap, as shown in Fig. 1 of ref. 2. The gastric arterial pressure was controlled by means of a roller pump. The venous pressure could be increased by appropriate clamping of the only vein connected to the preparation. It was measured in a small vein close to the stomach. The rectal temperature of the dog as well as the temperature of the mucosal chamber (Fig. 2) was controlled at 38’C. Dextrose (5% in saline) or saline was injected intravenously throughout. Sometimes blood was transfused from a donor animal. All injections were done intra-arterially either through the plastic tube of the extracorporeal circulatory system, or into a small branch of the artery which supplied the mucosal flap. Histamine and acetylcholine were usually given at a continuous rate controlled by an infusion pump. Since the ICP varies continuously, its mean value during a period was evaluated with a planimeter (20% of the experiments) or by a Grass integrator. Figures 4-6, 10, and 11 show records obtained with the latter procedure: each vertical line represents a certain area (cm*) depending on the paper velocity and sensitivity of the integration system. The experiments were divided into 15-min periods. To measure the amount of secretion, the gastric contents were
I) Sensitivity of recording system. Since we have not found reports on interstitial pressure measured with Guyton’s capsules of organs like the stomach which move frequently, we will describe in some detail the most important characteristics of our records. A displacement of 2 cm of the polygraph pen per 10 mmHg of pressure change at the ICP transducer was the usual sensitivity of the recording system. If the capsule was permeable, respiration artifacts, arterial pulse, and gastric peristaltic waves were recorded. Arterial pulse is shown in Fig. 3. The thickness of the ICP line is partly due to pulse waves and partly to respiratory artifacts as it was determined using a faster recording velocity. In the records of the middle section of the same figure, which were performed with a lower sensitivity, the pulse waves appeared only when the intragastric pressure was raised. This might be interpreted as an increase of the displacement of the arterial walls with each pulse wave, i.e., decreased “stiffness,” as a consequence of the augmented external pressure. Tension
FIG. 2. Diagram explanation).
of chamber
used
in experiments
transducer
(see text
for
FIG. 3. Correlation between intragastric pressure and ICP. Intragastric pressure was increased by introduction of a small amount of fluid into chamber, which was closed, with backplates (see Fig. 2) and filled with saline. Capsule implanted 12 days previously.
Downloaded from www.physiology.org/journal/ajplegacy at Glasgow Univ Lib (130.209.006.061) on February 13, 2019.
1416
ALTAMIRANO,
Figure 3 also illustrates the results obtained as the intragastric pressure was increased by introduction of small volumes of fluid into the chamber (the chamber was closed with backplates and filled with saline) : the intragastric pressure and the ICP varied by exactly the same amount. With appropriate amplification, intragastric changes of pressure amounting to a fraction of 1 mmHg were accurately recorded. 2) Effect of muscle activity on KY? As shown in Fig. 4, contraction of the gastric muscles increased the intragastric pressure and the ICP simultaneously if the chamber was Vetw.;
pfe9ssure
10 oc
lntrogastrtc pressure m m Hq
100 50 0
lntracapsular pressure m m Hq
too
/ i / 1, i
11
1.
1,
1 > 11
3
1 I
Ei
I 3
50 0 I1,1yl I
lntegratlon
1 Time
In mln
(large
1
partltlons)
FIG. 4. EfIect of gastric contractions on intragastric pressure and ICP. Chamber arranged as in Fig. 3. At signal I, prostigmine (50 pg) was injected intra-arterially. For “integration,” see METHODS. Capsule implanted 12 days previously. Venous pressure mmHg
Gastric crterial pressure mm %I
10 Cl
[I
200
’
, o.
m
i -
0 1 +I5
+5 Intracapsukx~ pressure mm Hg - 5
-15L
,
Integration
llIl~!il!
I II Time
l!lJ!I1IIpIy1ll
]11111111111!1!11!~
I in min
(large
partitions)
5. Effect of intra-arterial infusion of acetylcholine. At signal I, 10 pg/min of acetylcholine were infused for rest of experiment. Capsule implanted 25 days previously. FIG.
Gastric venous pressure mm Hg
IO oE-ry-
-
.
-y
*
REQUENA,
AND
PEREZ
closed and filled. At signal 1 of Fig. 4, Prostigmin was injetted intra-arterially. A temporary elevation of the muscle ‘tone augmented the basal intragastric pressure and the ICP to about 50 mmHg. Strong rhythmic contractions further increased both pressures another 40 mmHg. A single intra-arterial injection of 5-10 lug of acetylcholine usually determined a short but strong contraction of the gastric muscles and a large simultaneous increase of the ICP, which often surpassed 30 mmHg. The continuous infusion of low doses of acetylcholine resulted in the appearance of rhythmic contractions or increased the existing peristaltic movements, as illustrated in Fig. 5. Although in this figure the mean ICP (see METHODS) augmented to 5.6 mmHg, there was a period in which increased peristalsis coincided with transient reduction of the ICP to - 15 mmHg or more. When the muscle hyperactivity finished, the ICP was the same as before. In other experiments, strong contractions resulted in a progressive depression of the ICP that lasted minutes. This is illustrated in Fig. 6. The infusion of acetylcholine produced a series of short-lasting outbursts of peristaltic activity interspersed with intervals of rest. Initially the mean ICP varied around - 1 mmHg. After the seventh outburst, the ICP was reduced to -5 mmHg, but tended to rise slowly during the rest intervals. Each succeeding group of peristaltic waves again lowered the ICP to -5 mmHg. On the other hand, the mean ICP rose to 2.3 and 3.2 mmHg, respectively, during the illustrated periods. It is clear that the mean ICP was augmented by the muscle activity and not by an enlarged capillary filtration. Continuous infusion of histamine often increased gastric peristalsis, but never as strongly as acetylcholine: the ICP 3-7 mmHg during each wave. Large changed, at most, doses of intra-arterial histamine augmented the mean ICP even in the absence of muscle activity. These results seem to be caused by enlarged capillary filtration and will be discussed in detail in another report. Spontaneous gastric contractions usually appeared as peristaltic waves that traveled over the isolated mucosa from the upper (cardial) to the lower region. These waves modified the ICP not more than 4-6 mmHg, that is, much less than the muscle activity generated by infusion of acetylcholine or prostigmine. Figure 7 illustrates spon-
3
Integration 25 I5 pressure mm H9 IntrocapsuJaf
’E 0 -5 20 IA
0 * A Time
0
J(JLL!L in
min
(large
1
partitions)
FIG. 6. Effect of intra-arterial infusion of acetylcholine. Two minutes before A, infusion of 10 pg/min of acetylcholine had started. Between A and B, 13 min elapsed during which 4 additional outbursts of peristaltic activity occurred. Zero level of venous pressure had drifted and was corrected. Zero level of ICP was checked twice, as shown by 0. Capsule implanted 32 days previously.
FIG. 7. Spontaneous peristaltic waves and ICP. This experiment was done without backplates of chamber (Fig. 2). Therefore, pressure recorded by upper line reproduces mean change of pressure of chamber’s contents caused by a moving peristaltic wave in a mucosa that could be distended in places and shortened in others. ICP, which measures pressure at a fixed point, does not reproduce all changes shown in upper line. Calibration of upper record refers only to relative changes. Capsule implanted 34 days previously.
Downloaded from www.physiology.org/journal/ajplegacy at Glasgow Univ Lib (130.209.006.061) on February 13, 2019.
INTERSTITIAL
Gastric arterial pressure mm Hg
lntracopsulor pressure mm Hg
FLUID
PRESSURE
IN
CANINE
GASTRIC
100
,LV c ++‘OF5 0 - 5
A
Time
in min
(large
1417
MUCOSA
partltlons)
8. Effect on ICP of various circulatory changes. Between I and 2, splenic artery was completely obliterated. First incident in this recording was a calibration of zero of arterial pressure. Following drugs were injected intra-arterially : in 3, 5 pg of epinephrine; in 4, 5 pg of norepinephrine, and in 5 and 6, 5 pg of isoproterenol. Capsule implanted 27 days previously.
FIG. signals
taneous peristaltic activity and the concomitant ICP. Since the chamber was used without backplates (Fig. Z), this experiment reproduces the conditions that should obtain in a stomach filled with fluid. The amplitude and frequency of the waves are similar to the activity described in human stomach as tvpe 1 (3). We might note, however, that before mounting the stomach in the chamber, both vagus nerves were severed. Short spontaneous outbursts of strong peristaltic waves all exthat lasted 30-60 s appeared occasionally in nearly periments. As shown in the case of acetylcholine (Fig. 5), the ICP increased during each contraction, but often it fell to - 15 mmHg or lower during the relaxation phase. Movements of the capsule with respect to the pressure transducer did not seem to be the cause of these low pres10 cm of vertical .t would sures. More than displacemen have been necessary to produce such effects, while the capsule hardly moved a few millimeters even during the stronges t contractions. Similar transient lowerings of the IFP had been recorded during movements with the wick methods (IS), although the response of the wick to transien ts seem rather sluggish ( 15). 3) Effects of circulatory changes on KY. Besides muscle activity, circulatory changes modify the gastric ICP. As shown in Fig. 8, 5 lug of epinephrine determined a short lasting reduction of the ICP to about -5 mmHg. Norepinephrine caused a similar effect (Fig. 8). In the same animal, the complete obliteration of the splenic artery during 1 min (Fig. 8) or longer (not illustrated) lowered the ICP to the same level. During complete obliteration of the artery, the intracapillary pressure should fall the same level as venous pressure, increasing reabsorption of interstitial fluid and thereby reducing the ICP. In the same animal, 10 /rg of epinephrine or norepinephrine caused twice as much lowering of the ICP as that produced by obliteration of the artery (not illustrated), suggesting that these drugs act on the ICP not only by vasoconstriction (19), but, probably, also through relaxation of the gastric muscle (5). In agreement with this hypothesis, intra-arterial injections of atropine in doses of lo-30 mg reduced the ICP (Fig. lo), but always less than the maximal reduction produced by epinephrine. The sudden increase of the ICP observed in Fig. 8 when the circulation was reestablished was due to the intense
reactive hyperaemia that exists in the stomach, as demonstrated by other experiments in which the blood flow was measured. Isoproterenol, which produces muscle relaxation (5) but also increases blood flow, in small doses did not change the ICP significantly. In Fig. 8, the first dose of isoproterenol lowered the ICP. This effect, however, might be ascribed to some epinephrine that probably remained inside the injection system, since the second dose of 10 /lg of the drug produced the usual slight change of the ICP. Significant lowering of the ICP (5-8 mmHg) was observed with large doses of isoproterenol, 200 pg or more, but then, a fall of the systemic arterial pressure was evident, the gastric blood flow increased 2 or 3 times, and the stomach’s muscles relaxed deeply. Short-lasting depressions of the ICP of about 1 min duration appeared spontaneously in many experiments without evidence of muscle activity. These depressions happened in slightly anesthetized animals and might have been due to sympathetic discharges. Scholander et al. (16) have described lowering of the subcutaneous IFP in nonanesthetized animals when prodded. 4) ICP and uenous pressur up.The spontaneous portal pressure in our preparation varied around 5 mmHg, which is slightly less than the reported normal pressure (6). Elevation of this pressure immediately raised the in tragastric pressure and the ICP if the chamber was closed and filled with like this, initially the fluid (Fig. 9, part A). I n experiments distension of the capacitance vessels increases the volume of the mucosa and, hence, the pressure inside the chamber ( 13, 14). Since the IFP rose concomitantly, the net capillary filtration probably was not modified, as demonstrated by the immediate return of all pressures to the control level as soon as the venous pressure was lowered. Different results were obtained with the chamber connected as in Fig. 2, i.e., arranged to measure the volume of the mucosa under constant intragastric pressure (Fig. 9, part B). Now the ICP was modified only after relatively that is, after a significant net large changes of volume, capillary filtration occurred. In part B of Fig. 9, an increase of 2 ml, representing about 4% change in weight of the mucosa, did not modify the ICP. Probably this change was again a consequence of replenishing the capacitance vessels.
pressure vems m m Hg
Intrapstric pressure
lntrocapsubr pressure
mm Ho
IO
o i[---+-@-
Time
In m m
(large
partitions)
I
FIG. 9. Increase of venous pressure and ICP. A: chamber arranged as in Fig. 3 (completely closed). All changes of pressure inside chamber caused by venous hypertension were recorded exactly in both pressure transducers. B: chamber arranged as in Fig. 2 to monitor volume changes of preparation. Increase of volume of mucosa produced by filling of capacitance vessels does not modify ICP. Capsule implanted 16 days previously.
Downloaded from www.physiology.org/journal/ajplegacy at Glasgow Univ Lib (130.209.006.061) on February 13, 2019.
1418
ALTAMIRANO,
systemic pfeuure
orterid
150
mm Hg
w
Gastric pressure
orteriol mm tig
00
Gastric preuure
vonow mm Hg
11.
0 2o
0Er
.I
Ctmge d mucod volume ml
REQUENA,
AND
Pl%REZ
..
FIG. 10. Changes of volume and ICP during venous hypertension for extended periods of time. Between arrows, sensitivity of volume measurements was decreased as shown by corresponding scale. Signal at right indicates intra-arterial injection of 5 pg of epinephrine. Capsule implanted 16 days previously.
lntogrotion Time
in
mln
(large
portltions)
Figure 10 illustrates another experiment done with, the same mucosa of Fig. 8. In this case the venous pressure was raised twice during long periods. Again the volume of the mucosa was monitored. Clearly, the ICP increased pari passu with the volume of the preparation for as long as the venous pressure remained elevated. At the end of the experiment, the ICP returned to the control value, although the volume of the preparation exceeded its original volume by about 1 ml. Evidence of a large compliance is shown during the second rise of the venous pressure in which the volume continued increasing at a rate close to 0.6 ml/min but the ICP remained constant. The coefficient of capillary filtration for the whole stomach might be evaluated in experiments such as that illustrated in Fig. 10. Between the arrows, the volume of the preparation increased at a rate of 0.6 ml/min, presumably by increased capillary filtration (13, 14). We might assume that about 85 % of the venous pressure rise appears as an increase of the intracapillary pressure which normally is close to 15 mmHg. These assumptions are thoroughly discussed in the report of Folkow et al. (4). Considering the weight of the mucosal flap used in the experiment of Fig. 10 (40 g), the coefficient of capillary filtration amounts to 0.031 ml/100 g tissue min mmHg, which is much lower than the values found in cat’s intestine (4), but about >$-36th of the mean coefficient found by Johnson (10) in the intestine of dogs. Systemic arterial pressure mm Hq
150
GUStiC
150r
50 I--------------------
.-.
g 1
Time
in min
Oars
partitions)
11. Effect of hypertonic mannitol on ICP. At signal I, 0.5 ml of anesthetic was injected intravenously. At 2 and 4, 10 ml of hypertonic mannitol (18a/,) were injected intra-arterially. At 3, 10 mg of atropine were injected intra-arterially. Note change in scale of ICP recording. Capsule implanted 26 days previously. FIG.
5) Efect of hypertonic solutions on KY? Hypertonic solutions of mannitol (18 %) injected intra-arterially lowered the ICP concomitantly with a large rise of the local blood flow. Figure 11 illustrates a typical result. The dilatation of the gastric blood vessels was large enough to transiently depress the gastric arterial pressure. Figure 11 also illustrates that 10 mg atropine injected intra-arterially lowered the ICP by about 4 mmHg, probably by muscle relaxation but did not prevent the effects of mannitol. This latter compound inhibited the gastric muscles, but obviously its fundamental action on the ICP did not depend on muscle relaxation. Guyton (7) obtained similar results on the ICP after injections of concentrated dextran. This solution also reduces the pressure measured by the wick method ( 12). 6) Absolute ualue of ICY? As of now, it is clear that the ICP measured depended on muscle and circulatory parameters which probably varied in different animals. What follows, therefore, must be evaluated in the light of these facts. In 34 experiments the ICP was measured upon opening the abdominal cavity without any other surgical procedure and again later after the mucosa was mounted in the chamber. The mean values and standard error of the mean were, respectively, 1.08 & 0.43 and 0.19 & 0.34 mrnHg. The small difference between the two groups has doubtful physiological significance. With the intact stomach, it was rather difficult to level the capsule with the pressure transducer: 1 cm of difference in height was possible. On the other hand, in the mounted preparation the leveling could be adjusted to 1 mm. We might conclude that the surgical maneuvers related to the mounting of the mucosa did not significantly modify the ICP. A histogram of the ICP obtained in 49 animals with the stomach placed in the chamber is shown in Fig. 12. Probably the highest values were caused by reaction around the capsule. For instance, the dog in which the ICP measured 7.1 mmHg had a capsule enveloped by a fibrous in thickness (normal: about 0.5 mm). layer of about 2 mm Heavy infiltration of polymorphonuclear cells was evident. The capsule had been implanted 32 days previously. Most of the ICP’s in the histogram of Fig. 12 cluster between -2 and 2 mmHg. The mean amounts to 0.53 mmHg k a.34 (SEM). N evertheless, the spontaneous ICP was negative in about 43 % of the mucosae. 7) Intracapsular fluid and ICP. In all our experiments, the Guyton’s capsules were washed and filled with saline by
Downloaded from www.physiology.org/journal/ajplegacy at Glasgow Univ Lib (130.209.006.061) on February 13, 2019.
INTERSTITIAL
FLUID
PRESSURE
IN
CANINE
GASTRIC
MUCOSA
means of a very thin tube introduced into the intracapsular cavity. Saline was replaced by isotonic glucose or by blood plasma of the same animal in three experiments. In two of them, the ICP was the same, regardless of the solution introduced into the cavity. In the other experiment, the ICP increased about 3 mm when the intracapsular fluid was changed from saline to plasma. This higher ICP was not due to the larger concentration of proteins of the plasma, - _ - 1 s . since it increased further, another 2 or 3 mm, when plasma was replaced by isotonic glucose. In summary, we do not have evidence that the ICP varies with the protein concentration of the intracapsular fluid ( 11, 20). DISCUSSION
The experirnen ts performed with a closed chamber (sections I and 2 of RESULTS) demonstrate that any pressure exerted on the gastric mucosa can be accurately measured by means of Guyton’s capsules. Results described in sections 3-5 indicate that the ICP obeys Starling’s law of capillary filtration. Some drugs (section 3) and hypertonic mannitol (section 5) influence ICP as would be expected if the ICP were measuring IFP. Finally, our findings are similar to those obtained by Guyton (7-9) in an extensive series of experiments done with subcutaneous or muscle implanted that the ICP of capsules. We might conclude, therefore, this report is an accurate estimate of the IFP of the gastric su bmucosa. The meaning of the absolute values of the measurements made with Guyton’s capsules has been the subject of a long debate because they are at variance with values found by other procedures (9, 11, 12, 15-l 8, 20). In our opinion, Guyton’s experiments (5, 9) demonstrate that the needle method does not measure accurately the normal IFP. Toh;bPduqssure found by th e wick procedure usually amounts -3 mmHg (12, 15, 17), but it can be lower (18); it is well known that the pressure measured by Guyton’s capsules varies around -6 mmHg (9). Whether this difference arises from a semipermeable membrane newly formed around the implanted capsules as claimed by some authors (11, 20), or by other technical reasons (15), is controversial. Snashall and Boother (17) suggest that wick gel rather than fluid. Of -pressure is related to interstitial interest to us is that our I CP measure men ts gave on the average larger values than those found in subcutan eous or muscle tissue by either the capsule or wick procedures. This discrepancy merits some discussion. If the tissue moves continuously, as the stomach does, it is difficult to reach a complete cicatriz ation of the operatory wound within short periods. In our opinion, at least 40 days are required to obtain adequate healing of the stomach; otherwise, even slight trauma can cause small hemorrhages over the gastric wound. Since the capsules must be small, long delays necessitate the use of intracapsular stems (see METHODS) to avoid the filling of the capsule by connective tissue. About three quarters of our experiments were done within periods shorter than 40 days. Also the feed ing of the dogs seems important to obtain fast he aling. A soft diet is advisable, but nearly all our dogs were fed hard pellets of commercial food of about 1 cm in diameter. As- mentioned by Guyton (7) and con-
1419
Number
6
of
5
dogs
4 3 2 I 0 -
-7-6-5-4-3-2-l Introcapsular
0 I pressure
2
3 4 5 mm Hg
6
7
8
FIG. 12. Histogram of ICP measured in 49 mucosae. Rule followed to draw columns was: lower limit < values included in a column 5 upper limit. Mean value is 0.53 & 0.34 mmHg (SEM).
firmed by us, inflamation around the capsules increases the ICP. Although edema was seldom seen, polynuclear cells and other inflamatory signs were frequently observed in our autopsy specimens. Another factor which modifies the gastric ICP to a variable magnitude is the muscle activity. Ample evidence has been presented in this paper showing that the tone of gastric muscles and the peristaltic waves modify the ICP. In fairness, the reported IFP of subcutaneous tissue should be compared with the correspondent pressure of completely relaxed stomach. In our experiments, atropine reduced the ICP by about 5 mmHg. However, the drug was usually injected at the end of the experiments in which different procedures like histamine or acetylcholine infusion, or venous compression, etc. had increased the ICP. We do not have reliable statistical data on the ICP of normal and completely relaxed stomachs. We might note, however, that a reduction (by 5 mmHg) of the mean ICP reported here would give us values between - 3 and -5 mmHg, i.e., close to the results obtained by both the wick (12, 15 17) and capsule (9) methods in subcutaneous tissue. The lowest ICP’s, that is - 15 mmHg or lower, were measured during the relaxation phase of large peristaltic waves (section 2). If these strongly negative ICP’s were not caused by undetected artifacts, then probably all the ICP’s here reported might have a variable component due to muscle tension, which did not seem to have been inhibited by even the largest doses of atropine (30 mg intra-arterially). Since it has been shown that the IFP tends to be higher in warmer tissues (17), we might note that the temperature of our preparation was kept at 38°C. It has been claimed that the low ICP measured by Guyton ‘s method is a consequen ce of a semipermeable membar ne newly formed around the cavity inside the capsule (11, 20). This membrane, being semipermeable to the proteins, determines differences of osmotic pressure which cause the apparent low IFP. In all our experiments, the recordings were done with capsules whose cavities were filled with saline. In two out of three experiments done to study this problem (section 7), the ICP was the same whether the capsule was filled with saline, isosmotic glucose, or blood plasma. It is- obvious that in hollow organs such as the intestine, ureters, bladder, etc., the IFP will be variable, like that of the stomach. The quantitative evaluation of Starling’s law of capillary filtration will therefore be rather difficult to perform during normal activity. Results similar-to those illustrated in Figs. 5 and 6 indi-
Downloaded from www.physiology.org/journal/ajplegacy at Glasgow Univ Lib (130.209.006.061) on February 13, 2019.
ALTAMIRANO,
1420 (9) cate that the lymphatic pump” suggested by Guyton muscle conseem .s to be a work in the stoma .ch : powerful tractions lower the ICP by expulsion of fluid from the lymphatics, thereby decreasing the IFP. Finally, it could be of value to mention that the implan ted capsules I night be a good method to measure the intraluminal pressure of the stomach or intestine. The capsules stay in a fixed place, do not hinder the movements of either the gastric walls or of the gastric contents, and of the measure accu rately the pressure the lumen organ.
REQUENA,
AND
PEREZ
The authors thank Dr. A. C. Guyton for the gift of some capsules and for his useful comments on this report; they also thank Mrs. Carmen Paiva for her expert technical assistance. This work was supported in part by CONICIT (Consejo National de Investigaciones Cientificas y Tecnologicas) and by the Consejo de Desarrollo Cientifico y Humanistico, Universidad Central de Venezuela. Present address of M. Altamirano: Dept. of Physiology and Biophysics, School of Medicine, University of Puerto Rico, San Juan, P. R. Received
for publication
24 March
1975.
REFERENCES 1. ALTAMIRANO, acetylcholine. 2.
3.
4.
5. 6. 7.
8. 9.
10.
11.
M.
Alkaline
secretion produced by intra-arterial 168 : 787-803, 1963. ALTAMIRANO, AND R. P. DURBIN. Effects of gastric arterial and venous pressures on gastric secretion in the dog. Am. J. Physiol. 227 : 152- 160, 1974. CODE, C. F., N. C. HIGHTOWER, JR., AND C. G. MORLOCK. Motility of the alimentary canal in man. Am. J. Med. 13: 328-351, 1952. FOLKOW, B., 0. LUNDGREEN, AND I. WALLENTIN. Studies on the relationship between flow resistance, capillary filtration coefficient and regional blood volume in the intestine of the cat. Acta Physiol. &and. 57 : 270-283, 1963. GOODMAN, L. S., AND A. GILMAN. The Pharmacological Basis of Therapeutics. New York: Macmillan, 1970, p. 492 and 499. GRAYSON, J., AND D, MENDEL. Physiology of the Sp/anchnic Circulahon. London : Arnold, 1965, p. 158. GUYTON, A. C. A concept of negative interstitial pressure based on pressures in implanted perforated capsules. Circulation Res. 12: 399-414, 1963. GUYTON, A. C. Interstitial fluid pressure. II. Pressure-volume curves of interstitial space. Circulation Res. 16 : 452-460, 1965. GUYTON, A. C., H. J. GRANGER, AND A. E. TAYLOR. Interstitial fluid pressure. Physiol. Reu. 51 : 527-563, 1971. JOHNSON, P. C. Myogenic nature of increase in intestinal vascular resistance with venous pressure elevation. Circulation Res. 7: 992-999, 1959. KIRSCH, K., W. RAFFENBEUL, AND H. ROEDEL. Untersuchungen zur Ursache des negativen interstitiallen Gewebsdruckes (Guyton Kapsel). Pfluceers Arch. 328 : 193-204, 197 1. J. Physiol., London M., M. REQUENA,
12. LADEGAARD-PEDERSEN, H. J. Measurement of the interstitial pressure in subcutaneous tissue in dogs. Circulation Res. 26: 765770, 1970. S. Comparative studies on the adrenergic neuro13. MELLANDER, hormonal control of resistance and capacitance blood vessels in the cat. Acta Physiol. &and. Suppl. 176, vol. 50, 1960. J. R., AND A. SOTO-RIVERA. Effective osmotic 14. PAPPENHEIMER, pressure of the plasma proteins and other quantities associated with the capillary circulation in the hind limbs of cats and dogs. Am. J. Physiol. 152: 471-491, 1948. 15. PRATHER, J. W., D. N. BOWERS, D. A. WARREL, AND B. W. ZWEIFACH. Comparison of capsule and wick techniques for measurement of interstitial fluid pressure. J. A/$. Physiol. 31 : 942-945, 1971. 16 . SCHOLANDER, P. F., A. R. HARGENS, AND S. L. MILLER. Negative pressure in the interstitial fluid of animals. Science 16 1 : 32 l-328, 1968. 17 . SNASHALL, P. D., AND F. A. BOOTHER. Interstitial gel swelling pressure in human subcutaneous tissue measured with a cotton 1, 1969. wick. Chin. Sci. Mol. Med. 46 : 241-25 S. B., J. E. MAGGERT, AND P. F. SCHOLANDER. Inter18* STOMME, stitial fluid pressure in terrestrial and semiterrestrial animals. J. Apjd. Physiol. 27 : 123- 126, 1969. 19. SWAN, K. G., AND D. G. REYNOLDS. Effects of intra-arterial catecholamine infusions on blood flow in the canine gut. Gastroenterology 61 : 863-871, 1971. C. Dynamics of transcapillary fluid exchange. 20. WIEDERHIELM, J. Gen.
Pysiol.
52 : 29-61,
1968.
Downloaded from www.physiology.org/journal/ajplegacy at Glasgow Univ Lib (130.209.006.061) on February 13, 2019.
JOURNAL
OF PHYSIOLOGY
Vol. 229, No. 5, November 1975.
Interstitial M.
Printed
in U.S.A.
fluid pressure in canine gastric
ALTAMIRANO,
Departamento
Caracas,
de Ciencias
M. REQUENA, AND TATIANA Fisiolbgicas, Escuela de A4edicina J. M.
interstitial
fluid
pressure;
gastric
muscle
activity
1963, GUYTON (7) introduced a new procedure to measure the interstitial fluid pressure (IFP). Small plastic capsules pierced by holes were implanted in a tissue a few weeks before the pressure of the fluid inside the capsule (ICP) was measured. With this method, the IFP of normal subcutaneous tissue and muscle appeared negative by about -7 mmHg. Later, the use of cotton strands to connect the interstitial fluid with the measuring device, i.e., the wick method, confirmed that at least in skin and muscle the IFP is usually negative (12, 15-18). Estimates of that pressure in the mucosa of hollow organs like the stomach, where muscle activity might modify the IFP, have not been published. In the stomach, furthermore, hydraulic gradients across the epithelium seem to cause some type of secretion ( 1, 2), which adds interest to the knowledge of the local IFP. To study this problem, we implanted in the submucosa of the corpus modified versions -of Guyton’s capsules (7). The intracapsular pressure (ICP) was taken as an accurate estimated of the IFP (9). In this report we will describe a number of tests carried out to evaluate the performance of the capsules used by us. The mean value of the ICP was significantly larger than that found by Guyton (9), but experimental factors might explain at least part of the difl’erence. For instance, the activity of the gastric muscles substantially modifies the ICP. It was also found that the
IN
Central
de Venezuela,
Venezuela
ALTAMIRANO, M., M. REQUENA, AND TATIANA C. PEREZ. Interstitialjluidpressure in canine gastric mucosa. Am. J. Physiol. 229(5) : 1414-l 420. 1975.~-Guyton’s capsules were implanted in the submucosa of the gastric corpus of dogs. The pressure of the fluid inside the capsule (ICP) was measured between 12 and 42 days later after mounting the piece of the corpus in a Plexiglas chamber. The capsule was always filled with saline. In two out of three experiments, the ICP did not change significantly when the saline was replaced by isotonic glucose or blood plasma. Changes of pressure exerted on the surface of the stomach were accurately monitored by the ICP recordings. Changes of circulation produced by compression of the artery or the vein connected to the piece of mucosa, or by intra-arterial injections of epinephrine, norepinephrine, isoproterenol, or hypertonic mannitol, modified the ICP as predicted by Starling’s law of capillary filtration. Spontaneous activity of the gastric muscles or the activity, which followed intra-arterial injections of acetylcholine, prostigmine, or histamine, changed the ICP significantly. Intra-arterial atropine usually decreased the ICP by 3-5 mmHg. The mean value of the ICP in 49 animals was 0.53 zt 0.34 mmHg (SEM); it was negative in 43y0 of the experiments.
gastric
C. P&REZ Vargas, Uniuersidad
mucosa
capsules monitored accurately stomach, without interference gastric con tents and/or gastric
the pressure inside the with the movements of the walls.
METHODS
Mongrel dogs of either sex weighing between 12 and 25 kg were used. A small Guyton capsule (9) was implanted under Pentothal anesthesia in the gastric corpus close to the great curvature. A number of different capsules were used: the most satisfactory model is illustrated in Fig. 1. A plastic stem of about 2 mm diam was introduced to the very end of the capsule through the tube shown in the picture. The tube itself measured about 4-6 cm in length. The stem was pulled out for the final experiment, assuring the presence of a cavity inside the capsule notwithstanding the complete healing of the operatory wound. Eventually, the capsule was connected to a small displacement pressure transducer (Micron Instruments, Inc.) through plastic tubing. The capsules were placed in the submucosa through a short incision of the muscle layer that was subsequently sutured. Within 3 wk the capsule became enveloped by a thin film of connective tissue, usually of less than 0.5 111111 of thickness. The upper part of Fig. 1 illustrates a section of fixed mucosa after the capsule was pulled out. The small stumps of tissue that appear inside the cavity penetrated into the capsule through the wall perforations and were connected to a layer of newly formed tissue applied to the capsule’s walls. Histological examination of 26 mucosae disclosed that the epithelium overlying the capsule was normal in most cases. In a few animals, it was somewhat thinner than the neighboring epithelium. In many animals the capsule perforated the epithelium and was lost or remained in the gastric cavity attached to the wall by the plastic tubing. The dogs were anesthetized with pentobarbital 12-42 days after the implantation of the capsules. A flap of the fundus was mounted in a Plexiglas chamber as described in a previous paper (2). The chamber used and illustrated in Fig. 2 permitted us to perform a variety of measurements. When completely filled with isotonic solutions, as shown in the figure, it worked as a plethysmograph: changes in volume of the mucosa resulted in movement of fluid into or from the tube which hung from the tension transducer. The corresponding variations in weight were recorded using a Grass polygraph. Changes of 0.1 g were accurately measured (see Fig. 10). In other experiments, the lateral tube of the chamber was connected to a pressure transducer,
1414
Downloaded from www.physiology.org/journal/ajplegacy at Glasgow Univ Lib (130.209.006.061) on February 13, 2019.
INTERSTITIAL
FLUID
PRESSURE
Il\r CANINE
GASTRIC
1415
MUCOSA
carefully aspirated at the end of each period into tared tubes and weighed again. The gastric contents were analyzed for Cl- (Chloridometer), Na+ and K+ (flame photometer), and H+ by titration of an aliquot to pH 7. Samples of plasma were also taken for Na+, K+, and Cl- determinations. The area of stretched mucosa was measured in each experiment as described previously (2). RESULTS
FIG. 1. A: section of gastric wall fixed with formaldehyde after pulling away implanted capsule. White arrows show thickness of mucosa over capsule. Layer of fibrous tissue around capsule measures about 0.5 mm in thickness. B: photograph of a capsule.
so that small changes of pressure inside the chamber could be recorded (see Figs. 3 and 4). Most of the experiments were performed without the plastic triangular plates that support the mucosa (Fig. 2) and with the chamber empty or containing fluid just sufficient to cover the mucosa. Systemic pressure was measured in a small artery of the hindlegs. In some experiments the gastric arterial pressure was modified by means of an extracorporeal circulatory system. One femoral artery was connected by a plastic tube to the artery which supplied the mucosal flap, as shown in Fig. 1 of ref. 2. The gastric arterial pressure was controlled by means of a roller pump. The venous pressure could be increased by appropriate clamping of the only vein connected to the preparation. It was measured in a small vein close to the stomach. The rectal temperature of the dog as well as the temperature of the mucosal chamber (Fig. 2) was controlled at 38’C. Dextrose (5% in saline) or saline was injected intravenously throughout. Sometimes blood was transfused from a donor animal. All injections were done intra-arterially either through the plastic tube of the extracorporeal circulatory system, or into a small branch of the artery which supplied the mucosal flap. Histamine and acetylcholine were usually given at a continuous rate controlled by an infusion pump. Since the ICP varies continuously, its mean value during a period was evaluated with a planimeter (20% of the experiments) or by a Grass integrator. Figures 4-6, 10, and 11 show records obtained with the latter procedure: each vertical line represents a certain area (cm*) depending on the paper velocity and sensitivity of the integration system. The experiments were divided into 15-min periods. To measure the amount of secretion, the gastric contents were
I) Sensitivity of recording system. Since we have not found reports on interstitial pressure measured with Guyton’s capsules of organs like the stomach which move frequently, we will describe in some detail the most important characteristics of our records. A displacement of 2 cm of the polygraph pen per 10 mmHg of pressure change at the ICP transducer was the usual sensitivity of the recording system. If the capsule was permeable, respiration artifacts, arterial pulse, and gastric peristaltic waves were recorded. Arterial pulse is shown in Fig. 3. The thickness of the ICP line is partly due to pulse waves and partly to respiratory artifacts as it was determined using a faster recording velocity. In the records of the middle section of the same figure, which were performed with a lower sensitivity, the pulse waves appeared only when the intragastric pressure was raised. This might be interpreted as an increase of the displacement of the arterial walls with each pulse wave, i.e., decreased “stiffness,” as a consequence of the augmented external pressure. Tension
FIG. 2. Diagram explanation).
of chamber
used
in experiments
transducer
(see text
for
FIG. 3. Correlation between intragastric pressure and ICP. Intragastric pressure was increased by introduction of a small amount of fluid into chamber, which was closed, with backplates (see Fig. 2) and filled with saline. Capsule implanted 12 days previously.
Downloaded from www.physiology.org/journal/ajplegacy at Glasgow Univ Lib (130.209.006.061) on February 13, 2019.
1416
ALTAMIRANO,
Figure 3 also illustrates the results obtained as the intragastric pressure was increased by introduction of small volumes of fluid into the chamber (the chamber was closed with backplates and filled with saline) : the intragastric pressure and the ICP varied by exactly the same amount. With appropriate amplification, intragastric changes of pressure amounting to a fraction of 1 mmHg were accurately recorded. 2) Effect of muscle activity on KY? As shown in Fig. 4, contraction of the gastric muscles increased the intragastric pressure and the ICP simultaneously if the chamber was Vetw.;
pfe9ssure
10 oc
lntrogastrtc pressure m m Hq
100 50 0
lntracapsular pressure m m Hq
too
/ i / 1, i
11
1.
1,
1 > 11
3
1 I
Ei
I 3
50 0 I1,1yl I
lntegratlon
1 Time
In mln
(large
1
partltlons)
FIG. 4. EfIect of gastric contractions on intragastric pressure and ICP. Chamber arranged as in Fig. 3. At signal I, prostigmine (50 pg) was injected intra-arterially. For “integration,” see METHODS. Capsule implanted 12 days previously. Venous pressure mmHg
Gastric crterial pressure mm %I
10 Cl
[I
200
’
, o.
m
i -
0 1 +I5
+5 Intracapsukx~ pressure mm Hg - 5
-15L
,
Integration
llIl~!il!
I II Time
l!lJ!I1IIpIy1ll
]11111111111!1!11!~
I in min
(large
partitions)
5. Effect of intra-arterial infusion of acetylcholine. At signal I, 10 pg/min of acetylcholine were infused for rest of experiment. Capsule implanted 25 days previously. FIG.
Gastric venous pressure mm Hg
IO oE-ry-
-
.
-y
*
REQUENA,
AND
PEREZ
closed and filled. At signal 1 of Fig. 4, Prostigmin was injetted intra-arterially. A temporary elevation of the muscle ‘tone augmented the basal intragastric pressure and the ICP to about 50 mmHg. Strong rhythmic contractions further increased both pressures another 40 mmHg. A single intra-arterial injection of 5-10 lug of acetylcholine usually determined a short but strong contraction of the gastric muscles and a large simultaneous increase of the ICP, which often surpassed 30 mmHg. The continuous infusion of low doses of acetylcholine resulted in the appearance of rhythmic contractions or increased the existing peristaltic movements, as illustrated in Fig. 5. Although in this figure the mean ICP (see METHODS) augmented to 5.6 mmHg, there was a period in which increased peristalsis coincided with transient reduction of the ICP to - 15 mmHg or more. When the muscle hyperactivity finished, the ICP was the same as before. In other experiments, strong contractions resulted in a progressive depression of the ICP that lasted minutes. This is illustrated in Fig. 6. The infusion of acetylcholine produced a series of short-lasting outbursts of peristaltic activity interspersed with intervals of rest. Initially the mean ICP varied around - 1 mmHg. After the seventh outburst, the ICP was reduced to -5 mmHg, but tended to rise slowly during the rest intervals. Each succeeding group of peristaltic waves again lowered the ICP to -5 mmHg. On the other hand, the mean ICP rose to 2.3 and 3.2 mmHg, respectively, during the illustrated periods. It is clear that the mean ICP was augmented by the muscle activity and not by an enlarged capillary filtration. Continuous infusion of histamine often increased gastric peristalsis, but never as strongly as acetylcholine: the ICP 3-7 mmHg during each wave. Large changed, at most, doses of intra-arterial histamine augmented the mean ICP even in the absence of muscle activity. These results seem to be caused by enlarged capillary filtration and will be discussed in detail in another report. Spontaneous gastric contractions usually appeared as peristaltic waves that traveled over the isolated mucosa from the upper (cardial) to the lower region. These waves modified the ICP not more than 4-6 mmHg, that is, much less than the muscle activity generated by infusion of acetylcholine or prostigmine. Figure 7 illustrates spon-
3
Integration 25 I5 pressure mm H9 IntrocapsuJaf
’E 0 -5 20 IA
0 * A Time
0
J(JLL!L in
min
(large
1
partitions)
FIG. 6. Effect of intra-arterial infusion of acetylcholine. Two minutes before A, infusion of 10 pg/min of acetylcholine had started. Between A and B, 13 min elapsed during which 4 additional outbursts of peristaltic activity occurred. Zero level of venous pressure had drifted and was corrected. Zero level of ICP was checked twice, as shown by 0. Capsule implanted 32 days previously.
FIG. 7. Spontaneous peristaltic waves and ICP. This experiment was done without backplates of chamber (Fig. 2). Therefore, pressure recorded by upper line reproduces mean change of pressure of chamber’s contents caused by a moving peristaltic wave in a mucosa that could be distended in places and shortened in others. ICP, which measures pressure at a fixed point, does not reproduce all changes shown in upper line. Calibration of upper record refers only to relative changes. Capsule implanted 34 days previously.
Downloaded from www.physiology.org/journal/ajplegacy at Glasgow Univ Lib (130.209.006.061) on February 13, 2019.
INTERSTITIAL
Gastric arterial pressure mm Hg
lntracopsulor pressure mm Hg
FLUID
PRESSURE
IN
CANINE
GASTRIC
100
,LV c ++‘OF5 0 - 5
A
Time
in min
(large
1417
MUCOSA
partltlons)
8. Effect on ICP of various circulatory changes. Between I and 2, splenic artery was completely obliterated. First incident in this recording was a calibration of zero of arterial pressure. Following drugs were injected intra-arterially : in 3, 5 pg of epinephrine; in 4, 5 pg of norepinephrine, and in 5 and 6, 5 pg of isoproterenol. Capsule implanted 27 days previously.
FIG. signals
taneous peristaltic activity and the concomitant ICP. Since the chamber was used without backplates (Fig. Z), this experiment reproduces the conditions that should obtain in a stomach filled with fluid. The amplitude and frequency of the waves are similar to the activity described in human stomach as tvpe 1 (3). We might note, however, that before mounting the stomach in the chamber, both vagus nerves were severed. Short spontaneous outbursts of strong peristaltic waves all exthat lasted 30-60 s appeared occasionally in nearly periments. As shown in the case of acetylcholine (Fig. 5), the ICP increased during each contraction, but often it fell to - 15 mmHg or lower during the relaxation phase. Movements of the capsule with respect to the pressure transducer did not seem to be the cause of these low pres10 cm of vertical .t would sures. More than displacemen have been necessary to produce such effects, while the capsule hardly moved a few millimeters even during the stronges t contractions. Similar transient lowerings of the IFP had been recorded during movements with the wick methods (IS), although the response of the wick to transien ts seem rather sluggish ( 15). 3) Effects of circulatory changes on KY. Besides muscle activity, circulatory changes modify the gastric ICP. As shown in Fig. 8, 5 lug of epinephrine determined a short lasting reduction of the ICP to about -5 mmHg. Norepinephrine caused a similar effect (Fig. 8). In the same animal, the complete obliteration of the splenic artery during 1 min (Fig. 8) or longer (not illustrated) lowered the ICP to the same level. During complete obliteration of the artery, the intracapillary pressure should fall the same level as venous pressure, increasing reabsorption of interstitial fluid and thereby reducing the ICP. In the same animal, 10 /rg of epinephrine or norepinephrine caused twice as much lowering of the ICP as that produced by obliteration of the artery (not illustrated), suggesting that these drugs act on the ICP not only by vasoconstriction (19), but, probably, also through relaxation of the gastric muscle (5). In agreement with this hypothesis, intra-arterial injections of atropine in doses of lo-30 mg reduced the ICP (Fig. lo), but always less than the maximal reduction produced by epinephrine. The sudden increase of the ICP observed in Fig. 8 when the circulation was reestablished was due to the intense
reactive hyperaemia that exists in the stomach, as demonstrated by other experiments in which the blood flow was measured. Isoproterenol, which produces muscle relaxation (5) but also increases blood flow, in small doses did not change the ICP significantly. In Fig. 8, the first dose of isoproterenol lowered the ICP. This effect, however, might be ascribed to some epinephrine that probably remained inside the injection system, since the second dose of 10 /lg of the drug produced the usual slight change of the ICP. Significant lowering of the ICP (5-8 mmHg) was observed with large doses of isoproterenol, 200 pg or more, but then, a fall of the systemic arterial pressure was evident, the gastric blood flow increased 2 or 3 times, and the stomach’s muscles relaxed deeply. Short-lasting depressions of the ICP of about 1 min duration appeared spontaneously in many experiments without evidence of muscle activity. These depressions happened in slightly anesthetized animals and might have been due to sympathetic discharges. Scholander et al. (16) have described lowering of the subcutaneous IFP in nonanesthetized animals when prodded. 4) ICP and uenous pressur up.The spontaneous portal pressure in our preparation varied around 5 mmHg, which is slightly less than the reported normal pressure (6). Elevation of this pressure immediately raised the in tragastric pressure and the ICP if the chamber was closed and filled with like this, initially the fluid (Fig. 9, part A). I n experiments distension of the capacitance vessels increases the volume of the mucosa and, hence, the pressure inside the chamber ( 13, 14). Since the IFP rose concomitantly, the net capillary filtration probably was not modified, as demonstrated by the immediate return of all pressures to the control level as soon as the venous pressure was lowered. Different results were obtained with the chamber connected as in Fig. 2, i.e., arranged to measure the volume of the mucosa under constant intragastric pressure (Fig. 9, part B). Now the ICP was modified only after relatively that is, after a significant net large changes of volume, capillary filtration occurred. In part B of Fig. 9, an increase of 2 ml, representing about 4% change in weight of the mucosa, did not modify the ICP. Probably this change was again a consequence of replenishing the capacitance vessels.
pressure vems m m Hg
Intrapstric pressure
lntrocapsubr pressure
mm Ho
IO
o i[---+-@-
Time
In m m
(large
partitions)
I
FIG. 9. Increase of venous pressure and ICP. A: chamber arranged as in Fig. 3 (completely closed). All changes of pressure inside chamber caused by venous hypertension were recorded exactly in both pressure transducers. B: chamber arranged as in Fig. 2 to monitor volume changes of preparation. Increase of volume of mucosa produced by filling of capacitance vessels does not modify ICP. Capsule implanted 16 days previously.
Downloaded from www.physiology.org/journal/ajplegacy at Glasgow Univ Lib (130.209.006.061) on February 13, 2019.
1418
ALTAMIRANO,
systemic pfeuure
orterid
150
mm Hg
w
Gastric pressure
orteriol mm tig
00
Gastric preuure
vonow mm Hg
11.
0 2o
0Er
.I
Ctmge d mucod volume ml
REQUENA,
AND
Pl%REZ
..
FIG. 10. Changes of volume and ICP during venous hypertension for extended periods of time. Between arrows, sensitivity of volume measurements was decreased as shown by corresponding scale. Signal at right indicates intra-arterial injection of 5 pg of epinephrine. Capsule implanted 16 days previously.
lntogrotion Time
in
mln
(large
portltions)
Figure 10 illustrates another experiment done with, the same mucosa of Fig. 8. In this case the venous pressure was raised twice during long periods. Again the volume of the mucosa was monitored. Clearly, the ICP increased pari passu with the volume of the preparation for as long as the venous pressure remained elevated. At the end of the experiment, the ICP returned to the control value, although the volume of the preparation exceeded its original volume by about 1 ml. Evidence of a large compliance is shown during the second rise of the venous pressure in which the volume continued increasing at a rate close to 0.6 ml/min but the ICP remained constant. The coefficient of capillary filtration for the whole stomach might be evaluated in experiments such as that illustrated in Fig. 10. Between the arrows, the volume of the preparation increased at a rate of 0.6 ml/min, presumably by increased capillary filtration (13, 14). We might assume that about 85 % of the venous pressure rise appears as an increase of the intracapillary pressure which normally is close to 15 mmHg. These assumptions are thoroughly discussed in the report of Folkow et al. (4). Considering the weight of the mucosal flap used in the experiment of Fig. 10 (40 g), the coefficient of capillary filtration amounts to 0.031 ml/100 g tissue min mmHg, which is much lower than the values found in cat’s intestine (4), but about >$-36th of the mean coefficient found by Johnson (10) in the intestine of dogs. Systemic arterial pressure mm Hq
150
GUStiC
150r
50 I--------------------
.-.
g 1
Time
in min
Oars
partitions)
11. Effect of hypertonic mannitol on ICP. At signal I, 0.5 ml of anesthetic was injected intravenously. At 2 and 4, 10 ml of hypertonic mannitol (18a/,) were injected intra-arterially. At 3, 10 mg of atropine were injected intra-arterially. Note change in scale of ICP recording. Capsule implanted 26 days previously. FIG.
5) Efect of hypertonic solutions on KY? Hypertonic solutions of mannitol (18 %) injected intra-arterially lowered the ICP concomitantly with a large rise of the local blood flow. Figure 11 illustrates a typical result. The dilatation of the gastric blood vessels was large enough to transiently depress the gastric arterial pressure. Figure 11 also illustrates that 10 mg atropine injected intra-arterially lowered the ICP by about 4 mmHg, probably by muscle relaxation but did not prevent the effects of mannitol. This latter compound inhibited the gastric muscles, but obviously its fundamental action on the ICP did not depend on muscle relaxation. Guyton (7) obtained similar results on the ICP after injections of concentrated dextran. This solution also reduces the pressure measured by the wick method ( 12). 6) Absolute ualue of ICY? As of now, it is clear that the ICP measured depended on muscle and circulatory parameters which probably varied in different animals. What follows, therefore, must be evaluated in the light of these facts. In 34 experiments the ICP was measured upon opening the abdominal cavity without any other surgical procedure and again later after the mucosa was mounted in the chamber. The mean values and standard error of the mean were, respectively, 1.08 & 0.43 and 0.19 & 0.34 mrnHg. The small difference between the two groups has doubtful physiological significance. With the intact stomach, it was rather difficult to level the capsule with the pressure transducer: 1 cm of difference in height was possible. On the other hand, in the mounted preparation the leveling could be adjusted to 1 mm. We might conclude that the surgical maneuvers related to the mounting of the mucosa did not significantly modify the ICP. A histogram of the ICP obtained in 49 animals with the stomach placed in the chamber is shown in Fig. 12. Probably the highest values were caused by reaction around the capsule. For instance, the dog in which the ICP measured 7.1 mmHg had a capsule enveloped by a fibrous in thickness (normal: about 0.5 mm). layer of about 2 mm Heavy infiltration of polymorphonuclear cells was evident. The capsule had been implanted 32 days previously. Most of the ICP’s in the histogram of Fig. 12 cluster between -2 and 2 mmHg. The mean amounts to 0.53 mmHg k a.34 (SEM). N evertheless, the spontaneous ICP was negative in about 43 % of the mucosae. 7) Intracapsular fluid and ICP. In all our experiments, the Guyton’s capsules were washed and filled with saline by
Downloaded from www.physiology.org/journal/ajplegacy at Glasgow Univ Lib (130.209.006.061) on February 13, 2019.
INTERSTITIAL
FLUID
PRESSURE
IN
CANINE
GASTRIC
MUCOSA
means of a very thin tube introduced into the intracapsular cavity. Saline was replaced by isotonic glucose or by blood plasma of the same animal in three experiments. In two of them, the ICP was the same, regardless of the solution introduced into the cavity. In the other experiment, the ICP increased about 3 mm when the intracapsular fluid was changed from saline to plasma. This higher ICP was not due to the larger concentration of proteins of the plasma, - _ - 1 s . since it increased further, another 2 or 3 mm, when plasma was replaced by isotonic glucose. In summary, we do not have evidence that the ICP varies with the protein concentration of the intracapsular fluid ( 11, 20). DISCUSSION
The experirnen ts performed with a closed chamber (sections I and 2 of RESULTS) demonstrate that any pressure exerted on the gastric mucosa can be accurately measured by means of Guyton’s capsules. Results described in sections 3-5 indicate that the ICP obeys Starling’s law of capillary filtration. Some drugs (section 3) and hypertonic mannitol (section 5) influence ICP as would be expected if the ICP were measuring IFP. Finally, our findings are similar to those obtained by Guyton (7-9) in an extensive series of experiments done with subcutaneous or muscle implanted that the ICP of capsules. We might conclude, therefore, this report is an accurate estimate of the IFP of the gastric su bmucosa. The meaning of the absolute values of the measurements made with Guyton’s capsules has been the subject of a long debate because they are at variance with values found by other procedures (9, 11, 12, 15-l 8, 20). In our opinion, Guyton’s experiments (5, 9) demonstrate that the needle method does not measure accurately the normal IFP. Toh;bPduqssure found by th e wick procedure usually amounts -3 mmHg (12, 15, 17), but it can be lower (18); it is well known that the pressure measured by Guyton’s capsules varies around -6 mmHg (9). Whether this difference arises from a semipermeable membrane newly formed around the implanted capsules as claimed by some authors (11, 20), or by other technical reasons (15), is controversial. Snashall and Boother (17) suggest that wick gel rather than fluid. Of -pressure is related to interstitial interest to us is that our I CP measure men ts gave on the average larger values than those found in subcutan eous or muscle tissue by either the capsule or wick procedures. This discrepancy merits some discussion. If the tissue moves continuously, as the stomach does, it is difficult to reach a complete cicatriz ation of the operatory wound within short periods. In our opinion, at least 40 days are required to obtain adequate healing of the stomach; otherwise, even slight trauma can cause small hemorrhages over the gastric wound. Since the capsules must be small, long delays necessitate the use of intracapsular stems (see METHODS) to avoid the filling of the capsule by connective tissue. About three quarters of our experiments were done within periods shorter than 40 days. Also the feed ing of the dogs seems important to obtain fast he aling. A soft diet is advisable, but nearly all our dogs were fed hard pellets of commercial food of about 1 cm in diameter. As- mentioned by Guyton (7) and con-
1419
Number
6
of
5
dogs
4 3 2 I 0 -
-7-6-5-4-3-2-l Introcapsular
0 I pressure
2
3 4 5 mm Hg
6
7
8
FIG. 12. Histogram of ICP measured in 49 mucosae. Rule followed to draw columns was: lower limit < values included in a column 5 upper limit. Mean value is 0.53 & 0.34 mmHg (SEM).
firmed by us, inflamation around the capsules increases the ICP. Although edema was seldom seen, polynuclear cells and other inflamatory signs were frequently observed in our autopsy specimens. Another factor which modifies the gastric ICP to a variable magnitude is the muscle activity. Ample evidence has been presented in this paper showing that the tone of gastric muscles and the peristaltic waves modify the ICP. In fairness, the reported IFP of subcutaneous tissue should be compared with the correspondent pressure of completely relaxed stomach. In our experiments, atropine reduced the ICP by about 5 mmHg. However, the drug was usually injected at the end of the experiments in which different procedures like histamine or acetylcholine infusion, or venous compression, etc. had increased the ICP. We do not have reliable statistical data on the ICP of normal and completely relaxed stomachs. We might note, however, that a reduction (by 5 mmHg) of the mean ICP reported here would give us values between - 3 and -5 mmHg, i.e., close to the results obtained by both the wick (12, 15 17) and capsule (9) methods in subcutaneous tissue. The lowest ICP’s, that is - 15 mmHg or lower, were measured during the relaxation phase of large peristaltic waves (section 2). If these strongly negative ICP’s were not caused by undetected artifacts, then probably all the ICP’s here reported might have a variable component due to muscle tension, which did not seem to have been inhibited by even the largest doses of atropine (30 mg intra-arterially). Since it has been shown that the IFP tends to be higher in warmer tissues (17), we might note that the temperature of our preparation was kept at 38°C. It has been claimed that the low ICP measured by Guyton ‘s method is a consequen ce of a semipermeable membar ne newly formed around the cavity inside the capsule (11, 20). This membrane, being semipermeable to the proteins, determines differences of osmotic pressure which cause the apparent low IFP. In all our experiments, the recordings were done with capsules whose cavities were filled with saline. In two out of three experiments done to study this problem (section 7), the ICP was the same whether the capsule was filled with saline, isosmotic glucose, or blood plasma. It is- obvious that in hollow organs such as the intestine, ureters, bladder, etc., the IFP will be variable, like that of the stomach. The quantitative evaluation of Starling’s law of capillary filtration will therefore be rather difficult to perform during normal activity. Results similar-to those illustrated in Figs. 5 and 6 indi-
Downloaded from www.physiology.org/journal/ajplegacy at Glasgow Univ Lib (130.209.006.061) on February 13, 2019.
ALTAMIRANO,
1420 (9) cate that the lymphatic pump” suggested by Guyton muscle conseem .s to be a work in the stoma .ch : powerful tractions lower the ICP by expulsion of fluid from the lymphatics, thereby decreasing the IFP. Finally, it could be of value to mention that the implan ted capsules I night be a good method to measure the intraluminal pressure of the stomach or intestine. The capsules stay in a fixed place, do not hinder the movements of either the gastric walls or of the gastric contents, and of the measure accu rately the pressure the lumen organ.
REQUENA,
AND
PEREZ
The authors thank Dr. A. C. Guyton for the gift of some capsules and for his useful comments on this report; they also thank Mrs. Carmen Paiva for her expert technical assistance. This work was supported in part by CONICIT (Consejo National de Investigaciones Cientificas y Tecnologicas) and by the Consejo de Desarrollo Cientifico y Humanistico, Universidad Central de Venezuela. Present address of M. Altamirano: Dept. of Physiology and Biophysics, School of Medicine, University of Puerto Rico, San Juan, P. R. Received
for publication
24 March
1975.
REFERENCES 1. ALTAMIRANO, acetylcholine. 2.
3.
4.
5. 6. 7.
8. 9.
10.
11.
M.
Alkaline
secretion produced by intra-arterial 168 : 787-803, 1963. ALTAMIRANO, AND R. P. DURBIN. Effects of gastric arterial and venous pressures on gastric secretion in the dog. Am. J. Physiol. 227 : 152- 160, 1974. CODE, C. F., N. C. HIGHTOWER, JR., AND C. G. MORLOCK. Motility of the alimentary canal in man. Am. J. Med. 13: 328-351, 1952. FOLKOW, B., 0. LUNDGREEN, AND I. WALLENTIN. Studies on the relationship between flow resistance, capillary filtration coefficient and regional blood volume in the intestine of the cat. Acta Physiol. &and. 57 : 270-283, 1963. GOODMAN, L. S., AND A. GILMAN. The Pharmacological Basis of Therapeutics. New York: Macmillan, 1970, p. 492 and 499. GRAYSON, J., AND D, MENDEL. Physiology of the Sp/anchnic Circulahon. London : Arnold, 1965, p. 158. GUYTON, A. C. A concept of negative interstitial pressure based on pressures in implanted perforated capsules. Circulation Res. 12: 399-414, 1963. GUYTON, A. C. Interstitial fluid pressure. II. Pressure-volume curves of interstitial space. Circulation Res. 16 : 452-460, 1965. GUYTON, A. C., H. J. GRANGER, AND A. E. TAYLOR. Interstitial fluid pressure. Physiol. Reu. 51 : 527-563, 1971. JOHNSON, P. C. Myogenic nature of increase in intestinal vascular resistance with venous pressure elevation. Circulation Res. 7: 992-999, 1959. KIRSCH, K., W. RAFFENBEUL, AND H. ROEDEL. Untersuchungen zur Ursache des negativen interstitiallen Gewebsdruckes (Guyton Kapsel). Pfluceers Arch. 328 : 193-204, 197 1. J. Physiol., London M., M. REQUENA,
12. LADEGAARD-PEDERSEN, H. J. Measurement of the interstitial pressure in subcutaneous tissue in dogs. Circulation Res. 26: 765770, 1970. S. Comparative studies on the adrenergic neuro13. MELLANDER, hormonal control of resistance and capacitance blood vessels in the cat. Acta Physiol. &and. Suppl. 176, vol. 50, 1960. J. R., AND A. SOTO-RIVERA. Effective osmotic 14. PAPPENHEIMER, pressure of the plasma proteins and other quantities associated with the capillary circulation in the hind limbs of cats and dogs. Am. J. Physiol. 152: 471-491, 1948. 15. PRATHER, J. W., D. N. BOWERS, D. A. WARREL, AND B. W. ZWEIFACH. Comparison of capsule and wick techniques for measurement of interstitial fluid pressure. J. A/$. Physiol. 31 : 942-945, 1971. 16 . SCHOLANDER, P. F., A. R. HARGENS, AND S. L. MILLER. Negative pressure in the interstitial fluid of animals. Science 16 1 : 32 l-328, 1968. 17 . SNASHALL, P. D., AND F. A. BOOTHER. Interstitial gel swelling pressure in human subcutaneous tissue measured with a cotton 1, 1969. wick. Chin. Sci. Mol. Med. 46 : 241-25 S. B., J. E. MAGGERT, AND P. F. SCHOLANDER. Inter18* STOMME, stitial fluid pressure in terrestrial and semiterrestrial animals. J. Apjd. Physiol. 27 : 123- 126, 1969. 19. SWAN, K. G., AND D. G. REYNOLDS. Effects of intra-arterial catecholamine infusions on blood flow in the canine gut. Gastroenterology 61 : 863-871, 1971. C. Dynamics of transcapillary fluid exchange. 20. WIEDERHIELM, J. Gen.
Pysiol.
52 : 29-61,
1968.
Downloaded from www.physiology.org/journal/ajplegacy at Glasgow Univ Lib (130.209.006.061) on February 13, 2019.
