AMERICAN
JOURNAL
Vol. 228, No. 5, May
OF PHYSIOLOGY 1975. Printed in U.S.A.
hemodynamic
Reversible filtration
rate
defect in glomerular
after ischemic
injury
TERRANCE M. DAUGHARTY AND BARRY of Iris F. Ueki, Geraldine FitzGerald, and Robert Kidney Plpiology
Research
Laboratory,
and Cardr’ouascular
Veterans
Administration
Research
Institute,
Hospital, Chiuersi~
DAUGHARTY, TERRANCE M., AND BARRY M. BRENNER.R~UHS~~Z~ hemodynamic defect in glomerular Jiltrutian rate after z’schemz’c +.uy. Am. J. Physiol. 228(5): 1436-1439. 1975.-In Wistar rats with surface glomeruli 3 h of nearly complete occlusion of the left renal artery resulted in uniform falls (~50 “/) in ipsilateral wholekidney glomerular filtration rate (GFR) and single-nephron GFR (SNGFR) and in urinary findings consistent with impaired water reabsorption. Since the fall in SNGFR was accompanied by a proportional fall in glomerular plasma flow rate (GPF), and since net ultrafiltration pressure at afferent and efferent ends of the glomerular capillary was unchanged from preischemic levels, these findings suggest that the fall in SNGFR was a hemodynamic consequence of the fall in GPF. To test this hypothesis, GPF was restored to preischemic levels by means of acute infusion of homologous, isoncotic plasma. GPF and SNGFR uniformly increased, on average to preischemic levels, whereas net ultrafiltration pressure at afferent and efferent ends of the glomerular capillary again remained essentially unchanged. These studies demonstrate that the fall in SNGFR in this model of ischemic injury is causally related to the accompanying fall in GPF. Of interest, despite this reversal of the defect in GFR, fractional water reabsorption remained impaired. This restoration of GFR but not reabsorption by plasma infusion abruptly converts this ischemic lesion from nondiuretic to diuretic.
glomerular filtration;
capillary; ischemia; capillary blood flow;
renal microcirculation; hemodynamics
M. BRENNER Surface)
ultra-
WE HAVE RECENTLY REPORTED that 3 h of nearly complete occlusion of one renal artery results in a reduction in ipsilateral glomerular filtration rate (GFR) of approximately 50 % (3). This decline in GFR, affecting whole kidney and surface nephrons proportionately, occurred in the absence of measured changes in mean arterial pressure (z), mean glomerular transcapillary hydraulic pressure difference (Z), or net ultrafiltration pressure at afferent or efierent ends of the glomerular capillary network. Single-nephron filtration fraction (SNFF) also remained unchanged, indicating that the fall in single-nephron GFR (SNGFR) was accompanied by a proportional decline in glomerular plasma flow rate (GPF). Reduction in GPF, occurring in absence of a concomitant decline in mean glomerular capillary hydraulic pressure (&), resulted from large and proportional increases in aflerent (R,) and efferent (RE) arteriolar resistances. Since GFR has been shown to be highly plasma-flow
San Francisco of California,
(With 941.21,
the Technical and Departments
San Francisco,
California
Assistance
of Medicine
and
94143
dependent (Z), the possibility suggested by the above findings is that the reduction in GFR after ischemic injury might be reversed by restoring GPF to normal, preischemic values. To test this possibility, restoration of GPF to preischemic levels was achieved by expansion of plasma volume with homologous donor plasma in rats after unilateral renal ischemia. METHODS
Experiments were performed on male and female Munich-Wistar rats weighing 170-370 g and allowed free access to water and a standard rat pellet diet. The eflects of 3 h of nearly complete unilateral renal arterial occlusion on bilateral whole-kidney GFR (n = 3 rats) and on ipsilateral SNGFR (14 rats) were ascertained with experimental methods exactly as described in detail previously (3). In addition, the determinants of gloxnerular ultrafiltration were evaluated in each period as described before (2, 3). Since the effects of arterial occlusion on these determinants have already been reported (3), these measurements were repeated in the preocclusion period in only three rats and in five to six rats in the subsequent postocclusion period. Thereafter, homologous donor plasma was infused intravenously at 0.075 ml/min, and, when an average of 1.25 % body wt had been given, measurements of wholekidney (3 rats), single-nephron (12-14 rats), and microvascular (5-6 rats) functions were repeated. Once again, analytical techniques and calculations were exactly as described previously (2, 3). RESULTS
The effects of 3 h of unilateral renal ischemia and subsequent plasma loading on SNGFR, GPF, and several other pertinent measures of glomerular function are summarized in Fig. 1. After release of renal arterial occlusion, values for SNGFR were uniformly lower than in the preclamp period, on average by 49 % & 3 (SE) (1’ < .OOl). Values for SNGFR among different nephrons in the same rat showed the same minor degree of scatter after ischemia as before ischemia, indicating, as we have noted previously (3), that th’ IS f orm of renal injury does not result in heterogeneity of nephron function. As shown in Fig. 1, SNFF remained essentially unchanged; hence the fall in SNGFR was accompanied by a proportional decline in GPF. PGC avera.ged 46 & 1 mmHg before and after arterial
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.059.095.115) on January 13, 2019.
REVERSIBLE
DEFECT
IN
GFR
AFTER
PRE-
POST-
ISCHEMIA
ISCHEMIA
RENAL
1437
ISCHEMIA
POST- ISCHEMIA PLASMA+
LOADING
50
GPF (nlhin)
GPF
90
40
60
30
30 0
wmin)
20
i
PRESSURE (mm Hgl
SNGFR
(14)
SNGFR
SNFF
.38 (12)
-39
(12)
.30
(141
IO
l 16)
3.
(6) 16)
l
I5
8ARTERIOLAR RESISTANCE ( I O’“~dyn~sm-5)
6 _
01
I
I
I
1. Summary of effects of 3 h of unilateral renal ischemia and subsequent plasma loading on several pertinent measures of surfacenephron and microvascular function. Circles, brackets, and numbers within parentheses denote means, standard errors, and number of rats studied, respectively. FIG.
occlusion. Proximal tubule hydraulic pressure (PT) averaged 12 & 1 and 13 & 1 mmHg, respectively (Y > 0.5). Accordingly, aP;, given by P,c - Pr, remained essentially unchanged after ischemia. Protein concentration at the afferent end (CA) of the glomerular capillary network (as inferred from the concentration measured in femoral arterial plasma) averaged 5.5 =t 0.5 g/ 100 ml and 5.2 & 0.1 prior to and after ischemia, respectively (P > 0.5).” Eflerent arteriolar protein concentration (C,) averaged 8.8 & 0.1 g/l 00 ml before and 8*4 & 0.4 after occlusion (1’ > 0.5). The corresponding values for colloid osmotic pressure at aflerent (7~) and efferent (rE) ends of the glomerular capillary network, estimated with the LandisPappenheimer equation (7), are given in Fig. 1. As shown, equality of K E and E, denoting achievement of filtration pressure equilibrium, obtained both before and after renal ischemia. net ultrafiltration pressure at afferent -The ?T*) and efferent (PUFF 1 = PGc PGC pr = (P UFA 1 Inspection of Fig. 1 reveals that values for TA and TE represent the means of three, five, and six determinations for the preischemia, postischemia, and postischemia + plasma loading periods, respectively. In contrast, values for SNIFF are shown for a larger group of determinations, averaging 12, 12, and 14 measurements in these same periods. The reason for expressing the data for only the smaller number of determinations in the middle panel of Fig. 1 is that these values for 7TA and rlrli: can then be compared directly with values of AP obtained in these same animals. For this reason, TA is shown in Fig. 1 to rise slightly in the postischemia period (based on a mean value of CA of 5.6 g/100 ml, n = 5), whereas for all 12 rats in which measurements of CA were obtained the value averaged 5.2 g/100 ml, as stated in the text.
- PT - nE) ends of the capillary network averaged 17 & 2 and - 3 & 3 mmHg before and I6 & 1 (P > 0.5) and - I & 2 mmHg (P > 0.5) after arterial occlusion. The decline in GPF, occurring in the absence of a measured decline in PGC, was the consequence of large and proportional increases in Rh and R,. These results, summarized in Fig. 1, together with the other values obtained prior to and after renal arterial occlusion, are quantitatively similar to those we reported previously (3). Plasma loading resulted in uniform and highly significant (P < 0.001) reductions in systemic arterial hematocrit, onaverageto39.2 +0.9m1/100ml,compared to51.8 ho.7 and 53.7 & 0.8 ml/100 ml for the periods prior to and immediately after renal arterial occlusion. No significant alteration in mean systemic arterial pressure occurred : it averaged 114 & 2, 114 & 2, and 115 & 2 mmHg for the three consecutive periods studied. Plasma loading uniformly resulted in increases in SNGFR, on average from I8 + 2 to 36 & 3 nl/min, I’ < .OOl (Fig. l), the latter very nearly equal to that obtained prior to ischemia (37 =tr 3, P > 0.5). After plasma loading SNIFF fell slightly but significantly, on average from 0.39 to 0.30 (P < .OOl ), indicating that GPF must have increased slightly more than in proportion to the rise in SNGFR. Mean values for P,, and PT after plasma loading (44 & 1 and 14 =I= 1 mmHg) were not altered significantly from values prior to or after ischemia. Therefore, the restoration of SNGFR occurred in the absence of a significant change in AP (31 & 1 mmHg, 1) > 0.2). After plasma loading, CA and C, averaged 5.7 & 0.2 and 7.9 & 0.1 g/l 00 ml, corresponding to values of r A and rITE given in Fig+ 1. As shown, equality of rE ands persisted after plasma loading, the ratio PJ~ averaging 1.02 & -05 (I’ > 0.5). The basis for the moderate decline in SNFF after plasma loading can now be readily appreciated. Given the existence of filtration pressure equilibrium, SNFF will be determined C, and aP, As discussed in detail by just two quantities, elsewhere (5), SNFF will decrease when either AP decreases, C, increases, or both. Although not significant statistically, a decline in Z and a rise in C, were found after plasma loading. PUFA and PuFE were not increased after plasrna loading, averaging 12 & 2 and -2 & 3 mrnHg, respectively. Restoration of GPF, in the absence of a rise resulted from uniform and proportional declines in I&, in Rh and RR, as shown in Fig. 1. The effects of 3 h of left renal ischemia and subsequent plasma loading on total GFR, urine flow, and urine/plasma inulin ((U/P),,) rRc t ios were determined for both kidneys in each of three rats (Fig. 2). After left renal arterial occlusion, ipsilateral GFR fell, on average, by 49 % (Fig. 2, tok pan& open symbols) while contralateral GFR (solid symbols) tended to remain unchanged. Despite these moderately large declines in ipsilateral GFR, no corresponding decline in urine flow rate obtained (Fig. 2, middle pnnel). Thus, in addition to reducing GFR, ischemia resulted in a large fall in ipsilateral fractional water reabsorption, as indicated by the low (U/P)Ifl ratios relative to paired values on the noninjured side (Fig. 2, bottom panel). These results for whole-kidney function are similar to those previously reported (3). Plasma Ioading uniformly aug-
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.059.095.115) on January 13, 2019.
1438
T.
-1
PRE-
PQST-
ISCHEMlh
ISCHEMIA
POSTPLASMA+
ISCHEMIA LOADING
GFR Iml/min)
I5 URINE FLOW (&l/mid
JO 5 I
I
I
FIG. 2. Effects of 3 h of left renal arterial occlusion and subsequent plasma loading on several measures of bilateral renal function. Each symbol represents mean of 2 or more determinations. Solid symbols denote right or nonclamped kidney; open circles denote left, occluded kidney. Similar symbols identify kidneys from a single animal,
mented GFR and urine flow and decreased (U/P),, ratios bilaterally. Of interest was the finding that, whereas GFR values for nonclamped kidneys rose from normal to supernormal levels (0.99 to 1.30 ml/min), GFR values on the injured side rose from subnormal to normal (0.46 to 0.80 ml/min). Despite this difference, however, urine flow after plasma loading was higher (on average by 54 %) on the clamped than on the nonclamped side. Accordingly, (U/P)r, ratios remained lower on the injured than on the contralateral side. Thus, restoration of GFR but not fractional water reabsorption by plasma loading converts this ischemic renal lesion from nondiuretic to diuretic. DISCUSSION
In this experimental model of acute, unilateral renal ischemia, we have confirmed our recent findings (3) that whole-kidney and surface-nephron filtration rates are reduced proportionately, on average by approximately 50 %, after 3 h of nearly complete occlusion of the ipsilateral renal artery. In further accord with our previous findings, these declines in SNGFR occurred in the absence of mean changes in AP, PGC, PT, 3, or net ultrafiltration pressures at afferent and efferent ends of the glomerular capillary network. Net ultrafiltration pressure declined essentially to zero by the efferent end of the capillary network both before and after ischemic injury, denoting achievement of filtration pressure equilibrium under both conditions. Concurrently, values for SNFF remained unchanged, indicating that the measured reductions in SNGFR were accompanied by proportional declines in GPF. When filtration pressure equilibrium is achieved, as was found to be the case both prior to and after ischemic injury, SNGFR is determined by just three factors: GPF, s, and ‘lrA (4, 5). Since E and rA both remained essentially unchanged after ischemia, the reduction in SNGFR must have resulted from the simu taneous .nd proportional decline in GPF, reaffirr ning our previou conclusion that
M.
DAUGHARTY
AN’D
B. M.
BRENNER
SNGFR is highly plasma-flow dependent (2). A thorough discussion of the mechanism whereby a decrease in GPF brings about a proportional decrease in SNGFR can be found elsewhere (3). The reduction in GPF, occurring in the absence of a concomitant decline in I&, resulted from highly significant and proportional increases in RA and RE. We have suggested previously (3) that these resistance changes play a fundamental role in the pathogenesis of this form of acute renal failure. Further experimental support for this view derives from the recent work of Venkatachalam and associates (IO), who demonstrated narrowing of afferent and efferent arteriolar lumina without endothelial cell swelling in rats with acute renal failure induced by intramuscular injection of glycerol. Sheehan and Davis (8) and more recently others (1, 6, 9, 11) have observed the persistence of impaired blood flow to brain, heart, and kidney after relief of temporary, but severe, arterial occlusion (the so-called no-reflow phenomenon). Several workers (6, 11) have found that capillary endothelial cells tend to swell, often markedly, after severe ischemia and have suggested that such swelling accounts, at least in part, for the persistence of impaired organ perfusion In support of this possibility, perfusion of ischemic organs with hypertonic mannitol solutions is believed to improve organ blood flow (6, 1 1 ), presumably by an effect of this hypertonic, relatively impermeant, solute to reduce postischemic cell swelling. In the present study, isoncotic plasma loading uniformly restored whole-kidney and single-nephron GFR, GPF, RA, and RE to or toward preischemic levels. These effects occurred in the absence of important changes in nP, TA, or net ultrafiltration pressures at afferent or efferent TE, ends of the glomerular capillary network. Since filtration pressure equilibrium persisted after plasma loading, and since aa and TA were essentially constant, the restoration of SNGFR to preischemic levels was the consequence solely of the restoration of GPF. These findings thus demonstrate that hemodynamic alterations in GPF are primarily responsible for the decline in SNGFR observed in this model of renal ischemia. This restoration of GPF and SNGFR, and the accompanying return of RA and RE to normal levels, occurred after infusion of isoncotic plasma and thus was not dependent on the osmotic action of some hypertonic solute. These findings suggest that the postischemic reductions in GPF and SNGFR, and the associated increases in RA and I& were the result of factor(s) other than endothelial cell swelling. In this regard, we previously were unable to demonstrate capillary endothelial cell swelling in this model of renal ischemia (3), a finding in agreement with that of others studying alternate models of acute renal failure in the rat (10; M. Kashgarian, personal communication). For these reasons, we prefer to ascribe these measured alterations in RA and RE to alterations in the resistance to blood flow offered by single afferent and efferent arterioles. It remains to be determined whether these arteriolar resistance changes reflect intrinsic variations in vasomotor tone (such as vasoconstriction due to local accumulation of toxic metabolites during periods of renal &hernia) and/or result from alterations in the rheological properties of blood (since isoncotic plasma as employed in this study and rnannitol infusions
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.059.095.115) on January 13, 2019.
REVERSIBLE
DEFECT
IN
GFR
AFTER
RENAL
1439
KCHEMXA
of blood used by others (6, I 1) result in hemodilution hematocrit). It remains to be determined, too, whether a similar high degree of plasma-flow dependence of SNGFR exists in other commonly employed experimental models of acute renal failure. These
studies
were
supported
by
funds
from
the
Veterans
Ad-
ministration 13888). Portions
(1073-01)
and
the
National
of Health
have been published in abstract 1973). Address reprint requests to: B. M. Brenner, Chief, Section, Veterans Administration Hospital, 4150 Clement Francisco, Calif. 9412 1+
Am.
of this work
Institutes
Sue. Nephrol.
Received
form
(AM (Abstr.
6 : 29,
for publication
16 August
Nephrology Street, San
1974.
REFERENCES 1. AMES,
A., III, R. L. WRIGHT, hf. KOWADA, J. M. THURSTON, G. MAJNO. Cerebral ischemia. II. The no-reflow phenomenon. Am. J, Pat/ml. 52 : 437-453, 1968. B. M., J. L. TROY, T. M. DAUGHARTY, W. M. DEEN, 2. BRENNER, AND C. R. ROBERTSON. Dynamics of glomerular ultrafiltration in the rat. II. Plasma-flow dependence of GFR. Am. J. Physiol. AND
223: 1184-1190, 1972. 3. DAUGHARTY, T. M., I. F. UEKX, P. F. MERGER, AND B. M. BRENNER. Dynamics of glomerular ultrafiltration in the rat. V. Response to ischemic injury. J. Clin. Invest. 53: 105-l 16, 1974. 4. DEEN, W. M., C. R. ROBERTSON, AND B. M. BRENNER. A model of glomerular ultrafiltration in the rat. Am. J. Physiol. 223: 1178-1183, 1972. 5. QEEN, W. M., C. R, ROBERTSON, AND B. M. BRENNER. Glomerular ultrafiltration. Federation Proc. 33 : 14-20, 1974. 6. FLORES, J*, D. R. DIBONA, C. H. BECK, AND A. LEAF. The role
7.
8. 9. 10.
11.
of cell swelling in ischemic renal damage and the protective effect of hypertonic solute. J. C&n. Invest. 51: 118-126, 1972. LANDIS, E. M., AND J. R. PAPPENHEIMER. Exchange of substances through the capillary walls. In: Handbook of Physiology. Circulation. Washington, D.C. : Am. Physiol. Sot., 1963, sect. 2, vol. II, chapt. 29, p. 961-1034. SHEEHAN, H. L., AND J. C. DAVIS. Renal ischaemia with failed reflow. J. Puthol. Bacterial. 78 : 105-120, 1959. SUMMERS, W. K., AND R. L. JAMISON. The no-reflow phenomenon in renal ischemia. Lab. Invest. 25 : 635-643, 1971. VENKATACHALAM, M. A., H. RENNKE, AND D. SANDSTROM. Preglomerular vasoconstriction in acute renal failure. C/in. Res. 22 : 548A, 1974. WJLLERSON, J. T., W* J. POWELL, JR., T. E. G~INEY, J. J. STARK, C. A, SANDERS, AND A. LEAF. Improvement in myocardial function and coronary blood flow in ischemic myocardium after mannitol. J. Clin. Invest. 5 1: 2989-2998, 1972.
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.059.095.115) on January 13, 2019.
JOURNAL
Vol. 228, No. 5, May
OF PHYSIOLOGY 1975. Printed in U.S.A.
hemodynamic
Reversible filtration
rate
defect in glomerular
after ischemic
injury
TERRANCE M. DAUGHARTY AND BARRY of Iris F. Ueki, Geraldine FitzGerald, and Robert Kidney Plpiology
Research
Laboratory,
and Cardr’ouascular
Veterans
Administration
Research
Institute,
Hospital, Chiuersi~
DAUGHARTY, TERRANCE M., AND BARRY M. BRENNER.R~UHS~~Z~ hemodynamic defect in glomerular Jiltrutian rate after z’schemz’c +.uy. Am. J. Physiol. 228(5): 1436-1439. 1975.-In Wistar rats with surface glomeruli 3 h of nearly complete occlusion of the left renal artery resulted in uniform falls (~50 “/) in ipsilateral wholekidney glomerular filtration rate (GFR) and single-nephron GFR (SNGFR) and in urinary findings consistent with impaired water reabsorption. Since the fall in SNGFR was accompanied by a proportional fall in glomerular plasma flow rate (GPF), and since net ultrafiltration pressure at afferent and efferent ends of the glomerular capillary was unchanged from preischemic levels, these findings suggest that the fall in SNGFR was a hemodynamic consequence of the fall in GPF. To test this hypothesis, GPF was restored to preischemic levels by means of acute infusion of homologous, isoncotic plasma. GPF and SNGFR uniformly increased, on average to preischemic levels, whereas net ultrafiltration pressure at afferent and efferent ends of the glomerular capillary again remained essentially unchanged. These studies demonstrate that the fall in SNGFR in this model of ischemic injury is causally related to the accompanying fall in GPF. Of interest, despite this reversal of the defect in GFR, fractional water reabsorption remained impaired. This restoration of GFR but not reabsorption by plasma infusion abruptly converts this ischemic lesion from nondiuretic to diuretic.
glomerular filtration;
capillary; ischemia; capillary blood flow;
renal microcirculation; hemodynamics
M. BRENNER Surface)
ultra-
WE HAVE RECENTLY REPORTED that 3 h of nearly complete occlusion of one renal artery results in a reduction in ipsilateral glomerular filtration rate (GFR) of approximately 50 % (3). This decline in GFR, affecting whole kidney and surface nephrons proportionately, occurred in the absence of measured changes in mean arterial pressure (z), mean glomerular transcapillary hydraulic pressure difference (Z), or net ultrafiltration pressure at afferent or efierent ends of the glomerular capillary network. Single-nephron filtration fraction (SNFF) also remained unchanged, indicating that the fall in single-nephron GFR (SNGFR) was accompanied by a proportional decline in glomerular plasma flow rate (GPF). Reduction in GPF, occurring in absence of a concomitant decline in mean glomerular capillary hydraulic pressure (&), resulted from large and proportional increases in aflerent (R,) and efferent (RE) arteriolar resistances. Since GFR has been shown to be highly plasma-flow
San Francisco of California,
(With 941.21,
the Technical and Departments
San Francisco,
California
Assistance
of Medicine
and
94143
dependent (Z), the possibility suggested by the above findings is that the reduction in GFR after ischemic injury might be reversed by restoring GPF to normal, preischemic values. To test this possibility, restoration of GPF to preischemic levels was achieved by expansion of plasma volume with homologous donor plasma in rats after unilateral renal ischemia. METHODS
Experiments were performed on male and female Munich-Wistar rats weighing 170-370 g and allowed free access to water and a standard rat pellet diet. The eflects of 3 h of nearly complete unilateral renal arterial occlusion on bilateral whole-kidney GFR (n = 3 rats) and on ipsilateral SNGFR (14 rats) were ascertained with experimental methods exactly as described in detail previously (3). In addition, the determinants of gloxnerular ultrafiltration were evaluated in each period as described before (2, 3). Since the effects of arterial occlusion on these determinants have already been reported (3), these measurements were repeated in the preocclusion period in only three rats and in five to six rats in the subsequent postocclusion period. Thereafter, homologous donor plasma was infused intravenously at 0.075 ml/min, and, when an average of 1.25 % body wt had been given, measurements of wholekidney (3 rats), single-nephron (12-14 rats), and microvascular (5-6 rats) functions were repeated. Once again, analytical techniques and calculations were exactly as described previously (2, 3). RESULTS
The effects of 3 h of unilateral renal ischemia and subsequent plasma loading on SNGFR, GPF, and several other pertinent measures of glomerular function are summarized in Fig. 1. After release of renal arterial occlusion, values for SNGFR were uniformly lower than in the preclamp period, on average by 49 % & 3 (SE) (1’ < .OOl). Values for SNGFR among different nephrons in the same rat showed the same minor degree of scatter after ischemia as before ischemia, indicating, as we have noted previously (3), that th’ IS f orm of renal injury does not result in heterogeneity of nephron function. As shown in Fig. 1, SNFF remained essentially unchanged; hence the fall in SNGFR was accompanied by a proportional decline in GPF. PGC avera.ged 46 & 1 mmHg before and after arterial
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.059.095.115) on January 13, 2019.
REVERSIBLE
DEFECT
IN
GFR
AFTER
PRE-
POST-
ISCHEMIA
ISCHEMIA
RENAL
1437
ISCHEMIA
POST- ISCHEMIA PLASMA+
LOADING
50
GPF (nlhin)
GPF
90
40
60
30
30 0
wmin)
20
i
PRESSURE (mm Hgl
SNGFR
(14)
SNGFR
SNFF
.38 (12)
-39
(12)
.30
(141
IO
l 16)
3.
(6) 16)
l
I5
8ARTERIOLAR RESISTANCE ( I O’“~dyn~sm-5)
6 _
01
I
I
I
1. Summary of effects of 3 h of unilateral renal ischemia and subsequent plasma loading on several pertinent measures of surfacenephron and microvascular function. Circles, brackets, and numbers within parentheses denote means, standard errors, and number of rats studied, respectively. FIG.
occlusion. Proximal tubule hydraulic pressure (PT) averaged 12 & 1 and 13 & 1 mmHg, respectively (Y > 0.5). Accordingly, aP;, given by P,c - Pr, remained essentially unchanged after ischemia. Protein concentration at the afferent end (CA) of the glomerular capillary network (as inferred from the concentration measured in femoral arterial plasma) averaged 5.5 =t 0.5 g/ 100 ml and 5.2 & 0.1 prior to and after ischemia, respectively (P > 0.5).” Eflerent arteriolar protein concentration (C,) averaged 8.8 & 0.1 g/l 00 ml before and 8*4 & 0.4 after occlusion (1’ > 0.5). The corresponding values for colloid osmotic pressure at aflerent (7~) and efferent (rE) ends of the glomerular capillary network, estimated with the LandisPappenheimer equation (7), are given in Fig. 1. As shown, equality of K E and E, denoting achievement of filtration pressure equilibrium, obtained both before and after renal ischemia. net ultrafiltration pressure at afferent -The ?T*) and efferent (PUFF 1 = PGc PGC pr = (P UFA 1 Inspection of Fig. 1 reveals that values for TA and TE represent the means of three, five, and six determinations for the preischemia, postischemia, and postischemia + plasma loading periods, respectively. In contrast, values for SNIFF are shown for a larger group of determinations, averaging 12, 12, and 14 measurements in these same periods. The reason for expressing the data for only the smaller number of determinations in the middle panel of Fig. 1 is that these values for 7TA and rlrli: can then be compared directly with values of AP obtained in these same animals. For this reason, TA is shown in Fig. 1 to rise slightly in the postischemia period (based on a mean value of CA of 5.6 g/100 ml, n = 5), whereas for all 12 rats in which measurements of CA were obtained the value averaged 5.2 g/100 ml, as stated in the text.
- PT - nE) ends of the capillary network averaged 17 & 2 and - 3 & 3 mmHg before and I6 & 1 (P > 0.5) and - I & 2 mmHg (P > 0.5) after arterial occlusion. The decline in GPF, occurring in the absence of a measured decline in PGC, was the consequence of large and proportional increases in Rh and R,. These results, summarized in Fig. 1, together with the other values obtained prior to and after renal arterial occlusion, are quantitatively similar to those we reported previously (3). Plasma loading resulted in uniform and highly significant (P < 0.001) reductions in systemic arterial hematocrit, onaverageto39.2 +0.9m1/100ml,compared to51.8 ho.7 and 53.7 & 0.8 ml/100 ml for the periods prior to and immediately after renal arterial occlusion. No significant alteration in mean systemic arterial pressure occurred : it averaged 114 & 2, 114 & 2, and 115 & 2 mmHg for the three consecutive periods studied. Plasma loading uniformly resulted in increases in SNGFR, on average from I8 + 2 to 36 & 3 nl/min, I’ < .OOl (Fig. l), the latter very nearly equal to that obtained prior to ischemia (37 =tr 3, P > 0.5). After plasma loading SNIFF fell slightly but significantly, on average from 0.39 to 0.30 (P < .OOl ), indicating that GPF must have increased slightly more than in proportion to the rise in SNGFR. Mean values for P,, and PT after plasma loading (44 & 1 and 14 =I= 1 mmHg) were not altered significantly from values prior to or after ischemia. Therefore, the restoration of SNGFR occurred in the absence of a significant change in AP (31 & 1 mmHg, 1) > 0.2). After plasma loading, CA and C, averaged 5.7 & 0.2 and 7.9 & 0.1 g/l 00 ml, corresponding to values of r A and rITE given in Fig+ 1. As shown, equality of rE ands persisted after plasma loading, the ratio PJ~ averaging 1.02 & -05 (I’ > 0.5). The basis for the moderate decline in SNFF after plasma loading can now be readily appreciated. Given the existence of filtration pressure equilibrium, SNFF will be determined C, and aP, As discussed in detail by just two quantities, elsewhere (5), SNFF will decrease when either AP decreases, C, increases, or both. Although not significant statistically, a decline in Z and a rise in C, were found after plasma loading. PUFA and PuFE were not increased after plasrna loading, averaging 12 & 2 and -2 & 3 mrnHg, respectively. Restoration of GPF, in the absence of a rise resulted from uniform and proportional declines in I&, in Rh and RR, as shown in Fig. 1. The effects of 3 h of left renal ischemia and subsequent plasma loading on total GFR, urine flow, and urine/plasma inulin ((U/P),,) rRc t ios were determined for both kidneys in each of three rats (Fig. 2). After left renal arterial occlusion, ipsilateral GFR fell, on average, by 49 % (Fig. 2, tok pan& open symbols) while contralateral GFR (solid symbols) tended to remain unchanged. Despite these moderately large declines in ipsilateral GFR, no corresponding decline in urine flow rate obtained (Fig. 2, middle pnnel). Thus, in addition to reducing GFR, ischemia resulted in a large fall in ipsilateral fractional water reabsorption, as indicated by the low (U/P)Ifl ratios relative to paired values on the noninjured side (Fig. 2, bottom panel). These results for whole-kidney function are similar to those previously reported (3). Plasma Ioading uniformly aug-
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.059.095.115) on January 13, 2019.
1438
T.
-1
PRE-
PQST-
ISCHEMlh
ISCHEMIA
POSTPLASMA+
ISCHEMIA LOADING
GFR Iml/min)
I5 URINE FLOW (&l/mid
JO 5 I
I
I
FIG. 2. Effects of 3 h of left renal arterial occlusion and subsequent plasma loading on several measures of bilateral renal function. Each symbol represents mean of 2 or more determinations. Solid symbols denote right or nonclamped kidney; open circles denote left, occluded kidney. Similar symbols identify kidneys from a single animal,
mented GFR and urine flow and decreased (U/P),, ratios bilaterally. Of interest was the finding that, whereas GFR values for nonclamped kidneys rose from normal to supernormal levels (0.99 to 1.30 ml/min), GFR values on the injured side rose from subnormal to normal (0.46 to 0.80 ml/min). Despite this difference, however, urine flow after plasma loading was higher (on average by 54 %) on the clamped than on the nonclamped side. Accordingly, (U/P)r, ratios remained lower on the injured than on the contralateral side. Thus, restoration of GFR but not fractional water reabsorption by plasma loading converts this ischemic renal lesion from nondiuretic to diuretic. DISCUSSION
In this experimental model of acute, unilateral renal ischemia, we have confirmed our recent findings (3) that whole-kidney and surface-nephron filtration rates are reduced proportionately, on average by approximately 50 %, after 3 h of nearly complete occlusion of the ipsilateral renal artery. In further accord with our previous findings, these declines in SNGFR occurred in the absence of mean changes in AP, PGC, PT, 3, or net ultrafiltration pressures at afferent and efferent ends of the glomerular capillary network. Net ultrafiltration pressure declined essentially to zero by the efferent end of the capillary network both before and after ischemic injury, denoting achievement of filtration pressure equilibrium under both conditions. Concurrently, values for SNFF remained unchanged, indicating that the measured reductions in SNGFR were accompanied by proportional declines in GPF. When filtration pressure equilibrium is achieved, as was found to be the case both prior to and after ischemic injury, SNGFR is determined by just three factors: GPF, s, and ‘lrA (4, 5). Since E and rA both remained essentially unchanged after ischemia, the reduction in SNGFR must have resulted from the simu taneous .nd proportional decline in GPF, reaffirr ning our previou conclusion that
M.
DAUGHARTY
AN’D
B. M.
BRENNER
SNGFR is highly plasma-flow dependent (2). A thorough discussion of the mechanism whereby a decrease in GPF brings about a proportional decrease in SNGFR can be found elsewhere (3). The reduction in GPF, occurring in the absence of a concomitant decline in I&, resulted from highly significant and proportional increases in RA and RE. We have suggested previously (3) that these resistance changes play a fundamental role in the pathogenesis of this form of acute renal failure. Further experimental support for this view derives from the recent work of Venkatachalam and associates (IO), who demonstrated narrowing of afferent and efferent arteriolar lumina without endothelial cell swelling in rats with acute renal failure induced by intramuscular injection of glycerol. Sheehan and Davis (8) and more recently others (1, 6, 9, 11) have observed the persistence of impaired blood flow to brain, heart, and kidney after relief of temporary, but severe, arterial occlusion (the so-called no-reflow phenomenon). Several workers (6, 11) have found that capillary endothelial cells tend to swell, often markedly, after severe ischemia and have suggested that such swelling accounts, at least in part, for the persistence of impaired organ perfusion In support of this possibility, perfusion of ischemic organs with hypertonic mannitol solutions is believed to improve organ blood flow (6, 1 1 ), presumably by an effect of this hypertonic, relatively impermeant, solute to reduce postischemic cell swelling. In the present study, isoncotic plasma loading uniformly restored whole-kidney and single-nephron GFR, GPF, RA, and RE to or toward preischemic levels. These effects occurred in the absence of important changes in nP, TA, or net ultrafiltration pressures at afferent or efferent TE, ends of the glomerular capillary network. Since filtration pressure equilibrium persisted after plasma loading, and since aa and TA were essentially constant, the restoration of SNGFR to preischemic levels was the consequence solely of the restoration of GPF. These findings thus demonstrate that hemodynamic alterations in GPF are primarily responsible for the decline in SNGFR observed in this model of renal ischemia. This restoration of GPF and SNGFR, and the accompanying return of RA and RE to normal levels, occurred after infusion of isoncotic plasma and thus was not dependent on the osmotic action of some hypertonic solute. These findings suggest that the postischemic reductions in GPF and SNGFR, and the associated increases in RA and I& were the result of factor(s) other than endothelial cell swelling. In this regard, we previously were unable to demonstrate capillary endothelial cell swelling in this model of renal ischemia (3), a finding in agreement with that of others studying alternate models of acute renal failure in the rat (10; M. Kashgarian, personal communication). For these reasons, we prefer to ascribe these measured alterations in RA and RE to alterations in the resistance to blood flow offered by single afferent and efferent arterioles. It remains to be determined whether these arteriolar resistance changes reflect intrinsic variations in vasomotor tone (such as vasoconstriction due to local accumulation of toxic metabolites during periods of renal &hernia) and/or result from alterations in the rheological properties of blood (since isoncotic plasma as employed in this study and rnannitol infusions
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.059.095.115) on January 13, 2019.
REVERSIBLE
DEFECT
IN
GFR
AFTER
RENAL
1439
KCHEMXA
of blood used by others (6, I 1) result in hemodilution hematocrit). It remains to be determined, too, whether a similar high degree of plasma-flow dependence of SNGFR exists in other commonly employed experimental models of acute renal failure. These
studies
were
supported
by
funds
from
the
Veterans
Ad-
ministration 13888). Portions
(1073-01)
and
the
National
of Health
have been published in abstract 1973). Address reprint requests to: B. M. Brenner, Chief, Section, Veterans Administration Hospital, 4150 Clement Francisco, Calif. 9412 1+
Am.
of this work
Institutes
Sue. Nephrol.
Received
form
(AM (Abstr.
6 : 29,
for publication
16 August
Nephrology Street, San
1974.
REFERENCES 1. AMES,
A., III, R. L. WRIGHT, hf. KOWADA, J. M. THURSTON, G. MAJNO. Cerebral ischemia. II. The no-reflow phenomenon. Am. J, Pat/ml. 52 : 437-453, 1968. B. M., J. L. TROY, T. M. DAUGHARTY, W. M. DEEN, 2. BRENNER, AND C. R. ROBERTSON. Dynamics of glomerular ultrafiltration in the rat. II. Plasma-flow dependence of GFR. Am. J. Physiol. AND
223: 1184-1190, 1972. 3. DAUGHARTY, T. M., I. F. UEKX, P. F. MERGER, AND B. M. BRENNER. Dynamics of glomerular ultrafiltration in the rat. V. Response to ischemic injury. J. Clin. Invest. 53: 105-l 16, 1974. 4. DEEN, W. M., C. R. ROBERTSON, AND B. M. BRENNER. A model of glomerular ultrafiltration in the rat. Am. J. Physiol. 223: 1178-1183, 1972. 5. QEEN, W. M., C. R, ROBERTSON, AND B. M. BRENNER. Glomerular ultrafiltration. Federation Proc. 33 : 14-20, 1974. 6. FLORES, J*, D. R. DIBONA, C. H. BECK, AND A. LEAF. The role
7.
8. 9. 10.
11.
of cell swelling in ischemic renal damage and the protective effect of hypertonic solute. J. C&n. Invest. 51: 118-126, 1972. LANDIS, E. M., AND J. R. PAPPENHEIMER. Exchange of substances through the capillary walls. In: Handbook of Physiology. Circulation. Washington, D.C. : Am. Physiol. Sot., 1963, sect. 2, vol. II, chapt. 29, p. 961-1034. SHEEHAN, H. L., AND J. C. DAVIS. Renal ischaemia with failed reflow. J. Puthol. Bacterial. 78 : 105-120, 1959. SUMMERS, W. K., AND R. L. JAMISON. The no-reflow phenomenon in renal ischemia. Lab. Invest. 25 : 635-643, 1971. VENKATACHALAM, M. A., H. RENNKE, AND D. SANDSTROM. Preglomerular vasoconstriction in acute renal failure. C/in. Res. 22 : 548A, 1974. WJLLERSON, J. T., W* J. POWELL, JR., T. E. G~INEY, J. J. STARK, C. A, SANDERS, AND A. LEAF. Improvement in myocardial function and coronary blood flow in ischemic myocardium after mannitol. J. Clin. Invest. 5 1: 2989-2998, 1972.
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.059.095.115) on January 13, 2019.
