ARTICLES
NEWS IN PHYSIOLOGICAL SCIENCES
The Renal Pelvis: Machinery That Concentrates
Urine in the Papilla
Terry M. Dwyer1 and Bodil Schmidt-Nielsen2
1
Department of Physiology and Biophysics and Center for Excellence in Cardiovascular-Renal Research, University of Mississippi Medical Center, Jackson,
Mississippi 30216-4505; and 2Department of Physiology, University of Florida, Gainesville, Florida 32608
S
urprising to many, the pacemaker for ureteral peristalsis lies
within the kidney, near the cortical-medullary junction (3).
As a consequence, the renal pelvis (in unipapillate mammals)
and the renal calyces (in multipapillate animals) rhythmically
contract, forcing blood and fluid from the renal papilla; when
the muscle relaxes, fluid and blood return (17). We suggest
that the kidney papilla works as a pump, through alternating
positive and negative pressures generated by the peristaltic
contractions of the pelvic wall. Water moves into the collecting duct cells as a result of the small positive hydrostatic pressure on the walls of the cells, which is generated by the peristaltic wave pushing the fluid through the collecting ducts.
Subsequently, fluid moves out of the cells as a result of the
negative pressure generated by the elastic forces that expand
the papilla during the rebound, and fluid is removed from the
interstitium by the vasa recta. This model deals most specifically with mammals that have a relatively long papilla. (The
highest degree of urinary concentration is found in mammals
with the longest papilla, where peristalsis is expected to have
the greatest effect; see Ref. 17).
The structure of the renal pelvis
Mammals and birds are the only vertebrates known to produce a concentrated urine by means of a renal medullary
countercurrent system (18). These two countercurrent systems,
however, exhibit functional and anatomic differences that
appear to be related to the fact that birds excrete uric acid,
whereas mammals excrete most of their waste nitrogen in the
form of urea. In bird kidneys, there is no urea accumulation in
the renal medulla and there is no space for the urine to contact the renal medulla, because the medulla is surrounded by
tight sheets of connective tissue. In mammalian kidneys, urea
accumulation in the renal medulla plays an important role in
the mechanism that concentrates the urine in the collecting
0886-1714/03 5.00 © 2003 Int. Union Physiol. Sci./Am. Physiol. Soc.
www.nips.org
Downloaded from http://physiologyonline.physiology.org/ by 10.220.33.6 on June 17, 2017
Two decades ago, Bodil Schmidt-Nielsen and Bruce Graves documented the rhythmic contractions
of the renal pelvis in a remarkable video, visually demonstrating how peristaltic waves empty the
papilla and how the subsequent elastic recoil draws water from the collecting duct and into the
tethered-open ascending vasa recta. Thus a periodic hydrostatic gradient generates an axial
osmotic gradient. This review recapitulates the video and offers a link to figures and scenes
digitized from the original tape.
ducts. Furthermore, the mammalian renal medulla is surrounded by a muscular funnel-shaped pelvic wall (Fig. 1A),
which leaves an elaborate urinary space between the renal
medulla and the inside of the pelvic wall.
At this point, clarification concerning terminology is necessary. In the kidney with many papillae, such as human kidney,
each papilla is surrounded by a funnel-shaped calyx. This corresponds to what we call the pelvis in the kidney with only
one papilla. In the human kidney, it is the compartment
between the calyces and the ureter that is called the pelvis.
This compartment is not present in kidneys with one papilla,
where the pelvis is a direct extension of the ureter.
A plexus of smooth muscle fibers is found in the wall of the
renal pelvis and calyces. Two distinct layers exist in unipapillate kidneys. The inner layer contains fibers that run in
assorted directions, insert near the site where the pelvic wall
joins with the base of the papilla, and are continuous with the
ureteral smooth muscle; the outer layer is richly innervated but
is more diffuse and covers only the renal pelvis, connecting in
places with the inner layer but finally ending abruptly at the
junction with the ureter (Fig. 2A; Ref. 7). Although the layers
of smooth muscle in the caliceal wall in multipapillate kidneys are not as distinct (Fig. 2B), the inner fibers tend to insert
near the juncture of the caliceal wall and the base of the
papilla, whereas the outer fibers extend further (the musculus
levator fornicis; Ref. 15), forming a thickened ring around the
tubules and vasa recta that course toward the papilla (the
Ringmuskel der Papille; Ref. 10). The smooth muscles continue as far as the connective tissue associated with the arcuate arteries and veins (peripyramidal fibers; Ref. 16). A coordinated peristaltic contraction of these smooth muscles can
thus exert a rhythmic pumping action on the entire renal
medulla, a function first proposed by Henle (10) as a mechanism to promote the emptying of those tubules located in the
renal papilla.
News Physiol Sci 18: 16, 2003;
10.1152/nips.1416.2002
1
Urodynamic events occur in the renal pelvis
FIGURE 1. The renal pelvis and its extensions. A: structures in the renal
pelvis. The mammalian renal medullary countercurrent system exists in a
papilla that is surrounded by a urinary space bounded by a muscular pelvic
wall (17). Reprinted from Kidney International, volume 22, pages 613625,
1982. B: retrograde cast of the renal pelvis in the spotted hyena (Crocuta crocuta). The urinary space may have elaborate extensions in animals whose
habitat enforces long periods of water restriction, interspersed with ready
access to water (Reprinted from Ref. 11).
The mammalian renal pelvis makes it possible for urine to
contact the epithelium covering the inner and outer renal
medulla. In cross-sections of unipapillate mammalian kidneys, the pelvis appears to surround the inner medulla only
and to have a simple funnel shape. Casts of the pelvic urinary
space, however, reveal the presence of a series of leaf-like
extensions (Fig. 1B). There is an enormous variation in pelvic
spaces, which range from the simple funnel shape to the most
elaborate of pelvic extensions. Secondary pouches and fornices are found in all types of mammalian kidney. They are
urinary spaces in which the urine can contact the very thin
epithelium covering outer medullary tissue (see Ref. 13). In a
simple cross-section schematic of the rodent kidney, the inner
medulla and papilla can be seen to be hanging within the
pelvic space. Adjacent to the renal medulla is the septum,
2
News Physiol Sci • Vol. 18 • February 2003 • www.nips.org
Downloaded from http://physiologyonline.physiology.org/ by 10.220.33.6 on June 17, 2017
which is continuous with the ureter. Each septum has a series
of spokes that radiate toward the cortex and form the fornices.
Between the spokes, the outer margin of the septum is free
and has a semilunar edge. Behind this edge the secondary
pouches extend between the septum and the outer medulla.
The fornices reach all the way to the cortex. Capillaries are
located directly under the thin epithelium, and it is likely that
exchange of solute and water takes place between these capillaries and the urine reaching the pelvic extensions. The elaborate fornices in the sand rat maximize the possibility for
exchange of water and solutes between the urine in the pelvis
and the capillaries in the outer medulla. However, there is no
evidence to suggest that the ability to concentrate the urine is
enhanced in the mammals with the most elaborate and complex pelvic extensions. Rather, these structures allow the animals to respond to water intoxication by rapidly flattening the
solute gradient along the renal papilla (17).
“The renal pelvis is not a rigid and motionless hollow space.
This fact has been known for over 55 years....” (15), and that
was written in 1940! The hamster is particularly well suited for
observation of urodynamic events in the renal pelvis because
it has a single, rather long papilla. After careful removal of several layers of fat and connective tissue, the muscular contractions of the peripelvic wall become visible, particularly when
the papilla is transilluminated by a fiberoptic light. In the hamster, the rate of peristaltic contractions is normally 13/min (17),
compared with 7/min in miniature swine (but 1213/min in
isolated swine kidney; Refs. 4 and 20), 0.3/min in dog (but
1317/min in isolated canine kidney; Refs. 3 and 6), and 2
3/min in humans (2). This peristalsis has a profound effect on
the size and shape of the papilla. As the wall contracts, the
papilla narrows. Its diameter is reduced by as much as 20%,
which means that the cross-sectional area is reduced by as
much as 36%. At the moment the peristaltic wave moves past
its tip, the papilla rebounds upward. The average movement is
~300 m (17).
Smooth muscle is found in the wall of the renal pelvis up to
the base of the papilla in both unipapillate and multipapillate
kidneys but is not found in the papilla itself (Fig. 2). The peristaltic contractions are controlled by a pacemaker situated in
the uppermost parts of the pelvis (9), which overrides the
slower rhythms intrinsic to sections of the pelvic wall and
ureter.
It is possible to study the movement of the urine in the collecting ducts and in the pelvic space surrounding the renal
papilla during the peristaltic contractions of the muscular
pelvic wall by using a technique introduced by Steinhausen in
1964 (19). When the dye Lissamine Green SF is injected intravenously, it quickly reaches the capillaries of the kidney and
is filtered by the glomeruli. It soon appears in the proximal
tubules on the cortical surface of the kidney. As it moves
through the proximal tubule convolutions, the color fades
from the first convolutions as the dye enters the loops of
Henle. The colored fluid then returns to the distal tubules.
Since there has been water reabsorption from the renal
tubules, the dye is more concentrated and appears darker. A
similar sequence is seen in the kidney papilla. The dye first
appears in the capillaries and the vasa recta. After it has been
filtered and moved through the proximal convolutions, it
appears in the loops of Henle. Gradually the dye is cleared
from the loops, and after it has moved through the distal
tubules it enters the collecting tubules and the collecting
ducts. If dye is infused continuously, the urine moving through
the collecting ducts will remain green.
into the pelvic extensions thus has a lower osmolality than the
papillary tissue. It has been shown that full pelvic refluxes,
under these circumstances, serve to reduce the osmolality and
urea concentration of the renal medulla. As a consequence,
pelvic refluxes may serve to shorten the time it takes for a
water diuresis to develop following a large intake of water.
This may be particularly useful for desert animals that drink
large amounts of water periodically and therefore have to be
able to dilute the urine promptly to avoid water intoxication
(17).
Urine can reflux into the renal pelvis?
Urine leaving the ducts of Bellini at the tip of the papilla will
generally flow directly down the ureter after each pelvic contraction. Occasionally, however, urine fans out around the tip
of the papilla, briefly bathing the lower 50 m with urine (17).
This pattern, which we call tip reflux, is seen during constant
or decreasing urine flow. When the urine flow rate is rapidly
increasing, however, a different flow pattern is seen. The urine
no longer flows directly down the ureter as it leaves the ducts
of Bellini. Instead the urine is swept up into the pelvis, where
it reaches all the fornices and secondary pouches. During the
next contraction, the urine is swept down over the papilla and
enters the ureter. This pattern we call full pelvic reflux.
Full pelvic refluxes are physiologically induced when the
rate of urine flow increases faster than 0.05 l/min2. They continue for several minutes after the flow rate is no longer
increasing. During rising urine flow, the urine becomes more
dilute and its osmolality decreases. The urine being swept up
Downloaded from http://physiologyonline.physiology.org/ by 10.220.33.6 on June 17, 2017
FIGURE 2. The muscles of the pelvic wall. A:
in the unipapillate rat kidney, the smooth
muscle fibers (stained with anti-smooth muscle actin) terminate at the insertion of the
pelvic wall 12 mm from the base of the
papilla (arrowheads). B: complex geometry
of the human calyx requires a more effective
arrangement of the peristaltic machinery
than is found in unipapillate kidneys. Unlike
the renal pelvis in the rat, the smooth muscle
fibers accompany the renal calyx to the base
of the papilla (small arrowheads) and then
extend to the band of connective tissue containing the arcuate vasculature (*). Thus, in
humans, the peristaltic contraction moves
from the peripyramidal fibers (16), originating at the level of the arcuate vessels, and
moves down the musculus levator fornicis
(15), past the Ringmuskel der Papille (the
thick ring of smooth muscle encircling the
base of each papilla; large arrowheads; Ref.
10), compressing the entire length of the
papilla and closing the urinary space (u),
before proceeding toward the renal pelvis
and ureter. The blood supply to the calyx and
uroepithelium derives from arteries that
branch off the interlobular arteries (a) and
terminate in the tip of the papilla, potentially
allowing for a nutrient supply that supplements the vasa recta (1). A paired interlobar
vein (v) courses along the sinus fat deposit (f).
Peristalsis transports urine as a bolus down the ureter
Anyone who has passed a kidney stone knows that the
ureter also contracts peristaltically. Urine entering the ureter is
propelled toward the bladder in boluses, the ureter being
closed in front of, and behind, each bolus. At low urine flow
rates, the boluses are short; at higher flow rates, the boluses
become longer but the linear velocity of the boluses is
unchanged. Thus the flow in the ureter is very different from
flow in a pipe with a fixed diameter, where the linear velocity
would change in direct proportion to the flow rate. As the
urine flow increases further, the lengths of the boluses continue to increase until the ureter is filled from one end to the
other (17).
Peristaltic forces concentrate urine
As the peristaltic wave moves down over the renal papilla,
News Physiol Sci • Vol. 18 • February 2003 • www.nips.org
3
Downloaded from http://physiologyonline.physiology.org/ by 10.220.33.6 on June 17, 2017
FIGURE 3. Structural changes during and after a peristaltic contraction. A: renal papilla during a peristaltic contraction. The renal papilla of a Syrian hamster was
snared at a location ~1 mm from its tip during the contraction phase of the pelvic peristalsis. The collecting ducts are closed (17). B: renal papilla between peristaltic contractions. Similarly, a papilla was snared during the relaxation phase, showing the collecting ducts to be open, the cells to be smaller in volume, and
spaces to be present between the cells of the collecting ducts (17). C: patent ducts of Bellini. Ducts of Bellini are generally patent, as shown by these in a rabbit papilla (tissue was pressure perfused and examined with a scanning electron microscope; Ref. 5). D: prolapsed ducts of Bellini. In obese rabbits, the ducts of
Bellini are often seen distended by tissue prolapsed from within the papilla (5). A and B are reprinted from Kidney International, volume 22, pages 613625,
1982. C and D are reprinted from Ref. 5, with permission.
the urine is pushed through the papillary collecting ducts as a
bolus. Light micrographs of the papilla fixed during contraction show that the collecting ducts are closed (Fig. 3A) but that
they are wide open when the pelvic wall is relaxed (Fig. 3B).
As with the ureter, the collecting ducts are normally empty
and closed behind the bolus. The bolus is short during low
urine flow rates; the collecting ducts remain empty until the
next bolus is pushed through. At low urine flow rates, they
may be empty as much as 95% of the time, with fluid being
in contact with the collecting duct epithelium only briefly. The
linear velocity of the peristaltic wave is 1.6 mm/s and equals
the velocity of the short bolus. At higher urine flow rates, the
bolus is longer because the velocity of the leading edge,
which is determined by the urine production, is lower than the
velocity of the peristaltic wave pushing the trailing edge; at the
highest flow rates, the collecting ducts are empty only when
the peristaltic wave moves over the papilla. Thus, paradoxically, the contact time is prolonged (or the average linear
velocity of the urine in the collecting ducts is slower) at higher
flow rates compared with low urine flow rates.
As the peristaltic contraction moves down the papilla, the
collecting duct fluid is pushed through the ducts at a velocity
greater than the velocity with which the fluid is formed, thus
increasing the hydrostatic pressure on the wall of the ducts.
During low rates of urine production, only half of the fluid in
the terminal collecting duct exits the papilla from the ducts of
4
News Physiol Sci • Vol. 18 • February 2003 • www.nips.org
Bellini. The remaining volume is absorbed into the collecting
duct epithelial cells, whose volumes are increased by approximately the volume of the fluid absorbed from the collecting
duct lumen during each peristaltic contraction. As the contraction passes, collecting ducts close behind the urine bolus
(17).
Ostia of the ducts of Bellini are normally patent (Fig. 3C) but
frequently appear to contain prolapsed medullary tissue in
kidneys from obese rabbits (Fig. 3D), obesity being characterized by a modest degree of arterial hypertension (+11 mmHg
in these rabbits and similar elevations in obese dogs and
humans), an elevated renal interstitial hydrostatic pressure (19
vs. 9 mmHg in obese vs. lean dogs), and a markedly expanded
inner medullary interstitium (in rabbit, dog, and humans; Ref.
5). Thus the expanded interstitium may provide a greater than
normal resistance to the peristaltic rush of fluid through the
collecting ducts, which increases the interstitial pressure in the
renal parenchyma and results in an increased arterial pressure
systemically as well as prolapse of medullary contents out the
ostia.
The loops of Henle are also partly emptied during the pelvic
contraction. This can be seen following a bolus injection of
Lissamine Green SF, when the fluid in the loops becomes temporarily green. Each contraction appears to squeeze the fluid
out of the loops, with some fluid being pushed retrograde and
the remainder toward the tip.
Downloaded from http://physiologyonline.physiology.org/ by 10.220.33.6 on June 17, 2017
The peristaltic contractions of the pelvic wall profoundly
affect the papillary blood flow. The papilla inside the pelvic
wall becomes pale as the wall contracts around it, with the
capillaries narrowing during each contraction. With close
observation under the microscope, it can be seen that the
blood flow periodically stops and then briefly moves retrograde in descending capillaries. Measurements have shown
that blood flow is stopped ~30% of the time. Light and electron micrographs from renal papillae fixed at the very end of
contraction have shown that capillaries, as well as the loops
of Henle and the collecting ducts, were all tightly closed,
whereas those from relaxed papillae were all open. This intermittent pattern of flow may help preserve the osmotic gradient
along the renal papilla, since a continuous flow would result
in increased removal of papillary solutes (17).
Papilla rebound moves water into the vasa recta
The peristaltic contraction both compresses and stretches
the papilla, causing it to become longer and narrower. During
the early relaxation period (1 s), the papilla rebounds to its
shorter and broader relaxed form. The collecting ducts initially
remain closed, but water moves out of the cells into the interstitium due to negative hydrostatic pressure generated by the
elastic properties of the interstitial matrix. Since the ascending
vasa recta are tethered to other structures, they are opened as
the tissue expands during rebound, permitting water to enter
(14). Soon blood returns to the vasa recta, first to the descending and then to the ascending vasa recta, pushing along the
column of water that had entered the vasa recta from the interstitium. At this time, tubular fluid also returns to the loops of
Henle.
In late relaxation (2 s), collecting ducts open as urine flows
into them from above. The remaining structures continue to
function normally, with the shape of the papilla unchanged,
the vasa recta open with blood flow, and the loops of Henle
open with fluid continuing to flow through them.
The pelvic wall is necessary for peristalsis
If the pelvic wall is resected, the blood flow in the capillaries is continuous and the flow in the loops of Henle and
through the collecting ducts is also uninterrupted. Similarly, if
an intact pelvis is paralyzed by local anesthetic, the ureteral
peristalsis becomes uncoupled from the pelvis and the pelvis
becomes paralyzed. The flows in the collecting ducts, loops of
Henle, and vasa recta are now continuous, as it is when the
pelvic wall is removed or paralyzed. Physiological findings
have shown that, when the pelvic wall is removed or paralyzed, the osmolality and sodium concentration of the papillary tissue decreases significantly (17).
Mechanisms that concentrate urine in the papilla
To concentrate the urine in the collecting ducts, water must
be removed from the collecting ducts in excess of solutes. Part
of this water removal is caused by the accumulation of solutes
in the papillary interstitium, a consequence of ion transport
FIGURE 4. The animation. An animation of the model proposed in this paper
is available as a supplement on the web (see acknowledgements for URL),
and two frames are included in this figure. A: viewed longitudinally, the
papilla can be seen to be squeezed empty of fluid behind the peristaltic wave
and to be extruded in front of the wave. B: in cross-section, water is seen to
move from the collecting ducts to the open ascending vasa recta during
rebound, when the interstitial hydrostatic pressure is negative.
that occurs in the outer medulla. Mathematical models, however, cannot explain the actual concentrations reached by
some animals (12, 13). The model presented here (Fig. 4) proposes that the hydrostatic pressure generated by the pelvic
wall peristalsis contributes to the removal of water from the
collecting duct lumen in four steps. First, the force of the peristaltic contraction moves the bolus of urine not only down the
collecting duct but also into the cells lining the duct (Fig. 4A).
Second, negative interstitial pressures that develop during
rebound tend to move water from the epithelial cells into the
interstitium. Third, the tethered ascending vasa recta tend to
open before other structures (Fig. 4B; Ref. 14), generating a
negative intravascular hydrostatic force and drawing interstitial fluid into the lumen. Fourth, the ascending vasa recta has
an unusually large hydraulic permeability and low reflection
coefficient to albumin (14), allowing for vascular uptake of
both large and small osmotically active particless. Together,
these four steps constitute a highly effective mechanism for the
News Physiol Sci • Vol. 18 • February 2003 • www.nips.org
5
A supplemental video, “Renal Pelvis” by Bruce Graves and Bodil SchmidtNielsen, is available online for browsing. It can be accessed through the APS
website (http://www.apsarchive.org/renalpelvis/index.htm). The text has been
transcribed, and figures or short clips can be downloaded by double-clicking
the icons. A DVD version is also available from The American Physiological
Society; contact [email protected] or (301) 634-7180 to order.
The video recordings could not have been made without the skill and
inventiveness of Bruce C. Graves. The immunohistochemistry of Fig. 2 was
provided by Dr. Steven Bigler and Maxine Crawford. We wish to thank Louis
Clark for executing the animations seen in the video, Bobby Anderson, Bill
Buhner, and Jean Hurst for assistance in preparing figures and computer files
for the supplemental website (see URL above), Dr. Christoph Klett for assistance in conprehending the German text, and Drs. Allen Cowley, Michael
Flessner, John Hall, and Michael Hughson for their comments on an earlier
version of this paper.
Our research was supported by grants from the National Institutes of
Health (HL-51971 and AM-15972). The video was produced at Mt. Desert
6
News Physiol Sci • Vol. 18 • February 2003 • www.nips.org
Island Biological Laboratory in association with Grant AM-15972 and constituted a basis for the 1994 Krogh Lecture.
References
1. Baker SBD. The blood supply of the renal papilla. Br J Urol 31: 5359,
1959.
2. Björk L and Nylén O. Cineradiographic investigations of contraction in
the normal upper urinary tract in man. Acta Radiol Diagn (Stockh) 12:
2533, 1972.
3. Constantinou CE. Renal pelvic pacemaker control of ureteral peristaltic
rate. Am J Physiol 226: 14131419, 1974.
4. Djurhuus JC, Frokjaer J, Jorgensen TM, Knudsen L, Pham T, and Constantinou CE. Regulation of renal pelvic pressure by diuresis and micturition.
Am J Physiol Regul Integr Comp Physiol 259: R637R644, 1990.
5. Dwyer TM, Bigler SA, Moore NA, Carroll JF, and Hall JE. The altered structure of renal papillary outflow tracts in obesity. Ultrastruct Pathol 24:
251257, 2000.
6. Gosling JA and Constantinou CE. The origin and propagation of upper
urinary tract contraction waves. A new in vitro methodology. Experientia
32: 266267, 1976.
7. Gosling JA and Dixon JS. Morphologic evidence that the renal calyx and
pelvis control ureteric activity in the rabbit. Am J Anat 130: 393408,
1971.
8. Gullans SR. Metabolic basis of solute transport. In: Brenner and Rector’s
The Kidney (6th ed.), edited by Brenner BM. Philadelphia: W. B. Saunders, 2000, chapt. 5, p. 215-246.
9. Hannappel J and Lutzeyer W. Pacemaker localization in the renal pelvis
of the unicalyceal kidney. In vitro study in the rabbit. Eur Urol 4: 192-194,
1978.
10. Henle J. Zur Anatomie der Niere. Abhandlungen der Königich
Gesellschaft der Wissenschaften zu Göttingen 10: 134, 1862.
11. Hyrtl J. Das Nierenbecken der Säugethiere und des Menschen. Akademie
Wissenschaftlichen Math Naturwiss Wien 31: 107140, 1870.
12. Knepper MA, Chou CL, and Layton HE. How is urine concentrated in the
renal inner medulla? Contrib Nephrol 102: 144160, 1993.
13. Knepper MA, Saidel GM, Hascall VC, and Dwyer TM. Solute concentration in the renal inner medulla: interstitial hyaluronan as a mechanoosmotic transducer. Am J Physiol Renal Fluid Electrolyte Physiol. In press.
14. MacPhee PJ and Michel CC. Fluid uptake from the renal medulla into the
ascending vasa recta in anesthetized rats. J Physiol 487: 169183, 1995.
15. Narath PA. The hydromechanics of the calyx renalis. J Urol 43: 145176,
1940.
16. Puigvert A. Calyceal urodynamics. Urol Int 30: 282296, 1975.
17. Schmidt-Nielsen B. Function of the renal pelvis. In: Comparative Physiology, edited by Kinne RKH, Kinne-Saffran E, and Beyenbach KW. New
York: Karger, 1990, vol. 2, p. 103-140.
18. Schmidt-Nielsen B and Mackay WC. Comparative physiology of electrolyte and water regulation, with emphasis on sodium, potassium, chloride, urea and osmotic pressure. In: Clinical Disorders of Fluid and Electrolyte Metabolism (3rd ed.), edited by Maxwell MH and Kleemanx CR.
New York: McGraw-Hill, 1980, p. 37-89.
19. Steinhausen M. In-Vivo Beobachtungen an der Nierenpapille von Goldhamstern nach intravenöser Lissamingrün-Injection. Pflügers Arch 279:
195-213, 1964.
20. Yamaguchi OA and Constantinou CE. Renal calyceal and pelvic contraction rhythms. Am J Physiol Regul Integr Comp Physiol 257: R788R795,
1989.
Downloaded from http://physiologyonline.physiology.org/ by 10.220.33.6 on June 17, 2017
movement of interstitial fluid out of the medulla for any given
hydrostatic or osmotic gradient.
The structure of the model is unlike other mechanisms that
have been proposed. Most importantly, it is not a continuous
function of time; instead, the physical consequences of the
renal pelvic contractions must be integrated over a complete
peristaltic cycle for the transport of water to occur. Second, the
thermodynamic energy is not derived from ATP or ionic gradients within the papilla. This is fortunate, because the tissue
would appear to have meager energy reserves: blood flow is
slow, the partial pressure of oxygen is low (<10 Torr), mitochondria are few, and most energy is derived from anaerobic
glycolysis (8). Instead, we propose that the thermodynamic
energy is derived from an external source, namely the contraction of smooth muscle in the richly vascularized wall of
the renal pelvis. Third, certain proposed mechanisms require
that the interstitial compartments have a fixed volume or that
the renal papilla have a low compliance (12, 14). In fact, the
renal papilla must be very compliant, having no capsule and
little collagen. Instead, we propose that periodic contractions
of the renal pelvic wall enforce a constant volume, rhythmically moving fluid from the collecting ducts into the ascending vasa recta.
This model does not deal explicitly with the loops of Henle,
although their contents are emptied by peristalsis much as the
collecting ducts. Neither does it deal with the transport of
solutes, nor the dynamic effect of the repeated movements of
pure water into the papillary interstitium at the mouths of the
aquaporin channels, a flux that may alter the physical-chemical nature of the proteins, hyaluronan, and proteoglycans that
fill that space. We anticipate that application of contemporary
techniques will begin to elucidate these features of renal urodynamic activity.