- Email: [email protected]
Drug-induced changes in renal hippurate clearance as a measure of renal blood flow
Kidney International, Vol. 48 (1995), pp. 1617—1623
Drug-induced changes in renal hippurate clearance as a measure
of renal blood flow C. ANTOINETFE VISSCHER, DICK DE ZEEUW, PAUL E. DE JONG, WIM J. SLUITER,
and ROEL M. HUISMAN Groningen Institute for Dnig Studies1 (GIDS), Department of Medicine, Divisions of Nephrology and Endocrinology, University Hospital, Groningen, The Netherlands
Drug-induced changes in renal hippurate clearance as a measure of renal blood flow. We studied the accuracy of the plasma 1311-hippurate clearance technique to monitor drug-induced changes in renal blood flow
(RBF) by comparing it to a flow probe technique in six conscious, chronically instrumented dogs. Placebo caused no change in RBF, either established by hippurate clearance (ERPF5) or by renal blood flow probe (RBFproh,,). Enalaprilate induced a rise in ERPFhIP and RBFPrObC (+26 5 and 44 12%), as did dopamine (+16 4 and +33 5%). Intravenous
infusion of norepinephrine induced a rise in ERPFhIP (+2 6%, NS) and
in RBFprob,, (+18 3%), as did nitroprusside (+14 4% and +13 6%,
may still be accurately measured using renal hippurate clearance without knowing Eh1P, presuming that the intervention does not change EhIP. Yet, vasoactive substances are reported to affect EhIP [2—41. In addition, changes in hematocrit may occur affecting the
relation between RBF and hippurate (plasma) clearance. Thus, (drug)-induced changes in renal hippurate clearance are commonly considered to be an imprecise indicator even of changes in RBF [5—7]. To our knowledge, however, no studies have directly
NS). Indomethacin induced a fall in ERPFhP (—8 2%) and in RBFprObc
compared the clearance technique with the gold standard, the
angiotensin II (—5 2, —7 1, and +5 2%, respectively). Hematocrit (Hct) was affected by dopamine, norepinephrine, and nitroprusside (+2 1, +6 1, and —6 2%, respectively). Drug-induced changes in ERPF5Ip
renal artery blood flow probe, in measuring drug-induced changes in RBF and evaluated the impact of changes in EhIP or hematocrit on the measurement. In the present study we therefore compared changes in directly measured RBF with changes in plasma hippurate clearance with and without correction for EhIP and/or hematocrit in chronically
(—7 3%, NS), as did angiotensin II (—19 1 and —26 3%). Renal hippurate extraction (EhI) was affected by enalaprilate, dopamine, and
correlated well with changes in RBFPr,b (r = 0.902, P < 0.01). Changes in E1 did not independently affect this relation, whereas changes in Hct did: RBF(% of baseline) = 1.529 X IIERPFSiP(% of baseline) + 1.296 X zHct(% of baseline). These data indicate that drug-induced changes in plasma hippurate clearance can, even when changes in renal hippurate extraction are unknown, be used as a reliable indicator of changes in renal blood flow if changes in hematocrit are taken into account.
instrumented conscious dogs. These changes were induced by different drugs and hormones which are of clinical importance (enalaprilate, dopamine, norepinephrine, nitroprusside, indomethacin, and angiotensin II).
The renal clearance of exogenously infused hippurate has
Methods Six male mongrel dogs (25 to 33 kg body wt) were trained to become accustomed to the experimental situation and were kept
hippurate. Some hippurate, however, appears to escape renal
Instrumentation was performed in two steps under general
frequently been used since the fifties [1] as a means to establish on a salt restricted diet of less than I g of salt per day (H/D renal blood flow (RBF) both in the clinical and the laboratory Prescription Diet, Hill's Pet Products, Topeka, KS, USA). setting. This application is based on the presumption that plasma Instrumentation passing the kidney is nearly completely and constantly cleared of
anaesthesia (induction with thiopental, maintenance anaesthesia [1]. Thus, renal hippurate clearance can only be used as an with nitrous oxide and isoflurane). First, the left renal artery was absolute measure of RBF when corrected for EhIP. In practice this exposed retroperitoneally through a flank incision and dissected is not often done since EhP determination requires an invasive from surrounding tissue. Subsequently, a transit-time flow probe catheterization of the renal vein. Despite this, acute or even (Transonic Systems, Ithaca, NY, USA) of appropriate size was chronic changes in RBF induced by drug or other interventions placed around the renal artery. Furthermore, Tygon catheters (Norton, Akron, OH, USA; o.d. 1.8 mm, i.d. 1.0 mm) were placed in the descending aorta and the right atrium through the omocervical artery and vein, respectively [8]. Second, after a period of about two weeks recovery, the left renal vein was exposed through GIDS is part of the research school, Groningen Utrecht Institute for a midline incision and the spermatic vein was ligated to prevent Drug Exploration (GUIDE) sampling of non-renal blood. A heparin-coated Tygon catheter Received for publication February 6, 1995 and in revised form June 2, 1995 was placed in the renal vein through and fixed onto the inferior Accepted for publication June 2, 1995 caval vein. All leads were tunnelled subcutaneously to a location high on the back of the animal, and covered by a jacket. The © 1995 by the International Society of Nephrology clearance, so that renal hippurate extraction (EhIP) is never 100%
1617
1618
Visscher et al: Hippurate clearance and renal blood flow
collected by bladder catheterization. After a priming dose, 1311
E
300
hippurate was infused at a rate of 12 ml hr1 according to the method described by Donker et al [11]. After a stabilization period of two hours, six half-hour renal clearance periods were
200
performed: three baseline measurements followed by three drug or placebo measurements (starting 1 hour from administration). To measure renal hippurate extraction, arterial and renal venous blood samples were always drawn in the same order within three minutes of each other. Blood samples were immediately centrifuged at 4°C (480 g) and plasma was removed immediately to prevent diffusion of hippurate from the red cells into plasma.
100
Before the baseline and after the experimental period arterial
pa)0
0.
U-
blood samples were drawn to establish hematocrit (Hct). During
the baseline as well as during the experimental period mean arterial pressure (MAP: Disposable Pressure Transducer, Cobe® Laboratories) and left renal blood flow (RBFprobe: renal artery probe) were continuously registered.
0
0
100
200
300
RBFVO mi/mm
Fig. 1. Convlation between left renal blood flow measured by flow probe (RBFP,,h) and established by renal venous outflow (RBF,,,) in two dogs (• and 0). The solid line indicates the line of identity.
Data analysis Renal vascular resistance was calculated as MAP divided by RBFprobe.
The extraction ratio of hippurate was calculated as the difference between arterial (AhP) and renal venous 1311-hippurate concentration (Vh,) divided by Ah)P:
Eh1 =
(A
—
VP)/AhP
position of the renal vein catheter was verified before each Renal clearance of '311-hippurate was calculated according the experiment by oxygen saturation measurement [9]. All catheters following formula:
were kept patent by refilling them daily with a fresh heparin
solution (2000 IE/ml). Flow probes were precalibrated by the producer. In two dogs under general anaesthesia calibration was verified in vivo after the final experiment. This, by comparing flow probe values with renal venous outflow measurements, Changes in renal blood flow were induced by bleeding: One flow probe constantly overestimated (22 7%) whereas the other flow probe slightly underestimated (—9
ERPFh1(U) =
unneblP
X urine flow rate/A
Plasma clearance of hippurate was calculated as follows:
ERPF(I) = infusate X infusion rate/A Renal blood flow (RBFhIP) was subsequently calculated as:
5%) renal blood flow. Furthermore, calibration curves were RBFhP = ERPFhP/{EIJP x (1 — Hct)] perfectly linear (r = 0.98 and >0.99; Fig. 1). By intra-arterially injected angiotensin II (5 to 10 j.tg) it was verified in vivo that no The means of the three clearance periods before and after drug or placebo administration were used as the respective baseline drifting of zero level occurred [10]. SEM. experimental values. Values are expressed as mean Experimental design
Statistical analysis was performed with the commercially available
After completion of the instrumentation the animal was allowed to recover for a least one week. The acute effects of placebo, enalaprilate (ACEi), dopamine (DOP), norepinephrine (NE), nitroprusside (NIP), indomethacin (IND), and angiotensin II (Ang II) were studied on subsequent experimental days in a randomized order. Placebo, ACEi (10 mg enalaprilate i.v.) and IND (1 mg kg' Indocid® P.D.A. i.v.; a gift of MSD, Haarlem, The Netherlands) were given in a single dosage, whereas DOP, NE, NIP, and Ang II were continuously infused (15 ml hr1 i.v.)
software program Instat (GraphPad Software, San Diego, CA, USA). Differences between baseline and experimental values during placebo were tested by the Student's paired t-test. Druginduced changes were expressed as percentage change from baseline, and subsequently tested by comparing them with the placebo values by the paired t-test. The relation between changes
at a dose titrated to a 10 to 30% change in mean arterial pressure and renal blood flow. Drugs were dissolved in 5% glucose. To
Results
prevent drug interaction or carry-over, the drugs were studied during at least one week intervals. Before the experimental day dogs had free access to water, but they were deprived of food for at least eight hours. During the experiment the trained dog was conscious and hanging quietly in a hammock. To provide adequate urine production 5% glucose was infused continuously at a rate of 250 ml hr'. Urine was
in ERPFhIP, EhiP, and Hct and changes in RBFPrObC was tested by
multiple variable regression analysis. Differences and correlations were considered significant at a level of 5%.
Baseline values
Table 1 shows the baseline values of all parameters established during the experimental days with placebo (N = 6), ACEi (N = 5), dopamine (N = 5), norepinephrine (N = 5), nitroprusside (N = 6), indomethacin (N = 6), and angiotensin II (N = 6). No significant differences were observed, although baseline values
varied slightly between the experimental days. Furthermore,
1619
J/isscher et a!: Hippurate clearance and renal blood flow
Table 1. Baseline values of mean arterial pressure (MAP), renal vascular resistance (RVR), left renal blood flow (RBFprObe), plasma [ERPFhIP (I)] and renal hippurate clearance [ERPFhP (U)I, renal hippurate extraction (Eh,P), haematocrit (Hct), and renal blood flow as established by plasma hippurate clearance (RBFhIP)
MAP mm Hg Plac
ACEi DOP NE NIP IND
Ang II
113 109 113 109 109 111 109
0.505 0.415
Hct
mi/mm
246 8
0.460 0.023 0.449 0.042 0.480 0.046
5 6 5 6 6 5 4
ERPFhIP (U)
ERPFhIP (I)
RBFprOb
RVR mm Hg/mm/mi
0.075 0.048
0.446 0.051 0.416 0.044
277
306 26
305 27 294 34 291 20 294 33 322 35 325 24 322 32
248 16 241 21 225 24 276 29 265 30 30
291
0.77
0.02
297 28 297 33 316 36 341 27 314 34
0.77 0.78 0.75 0.74
0.02 0.02 0.03 0.02
mI/mm
39 3 42 1 40 3 39 1 40 2 38 2 40 2
0.77 0.02 0.75 0.02
34
RBFhP
%
EhI
618 43 655 59 639 27 646 80 693 67 701 46 726 62
Values shown during placebo (Plac), enalaprilate (ACEi), dopamine (DOP), norepinephrine (NE), nitroprusside (NIP), indomethacin (IND), and angiotensin II (Ang II) experiments.
Placebo
IND
NIP
NE
DOP
ACEI
Ang II
—
400 E .0 0 0.
U-
200 -
400 E
0.
0
-
U-
200 -
E *#
'I
0.80 0.
*
w 0.60 -
+
—
+
—
+
+
—
—
+
—
+
—
+
Fig. 2. Individual values of left renal blood flow measured by flow probe (RBFP,.c,b), plasma hippurate clearance (ERPFh,I,), and renal hippurate e.rtraction ratio (E1,q,), before and after administration of placebo, enalaprilate (ACEi), dopamine (DOP), norepmnephrine (NE), nitroprusside (NIP), indomethacin
(IND), or angiotensin II (Ang II) (*P < 0.05 vs. baseline, #P < 0.05 vs. placebo).
baseline values of ERPFhIP(I) and ERPFhiP(U) were comparable.
This indicates that at steady state the appearance rate of hippurate in the plasma equals the appearance rate in the urine, and
Drug-induced changes
Figure 2 shows the individual values of RBFPrOBe, ERPFhIP, and
that ERPFhP(I) can be substituted by ERPFhIP(U). Since EhIP on the different experimental days, whereas in Figure 3 the ERPFhP(I) is not subject to errors introduced by incomplete average placebo corrected changes are shown. urine collection or dead-space, this entity was used to establish drug-induced changes in renal hippurate clearance.
Placebo. Placebo infusion did not induce significant changes in MAP and RVR (+2 1 and +3 2%). As can be seen in Figure
1620
Visscher et a!: Hippurate clearance and renal blood flow
A
Nitroprusside. Nitroprusside was given at a mean dosage of 0.29 0.03 mg kg1 - hr'. This resulted in a fall in MAP as well as RVR (—23 2 and —31 6%). RBFprobe was not significantly affected, whereas ERPFhIP rose significantly (Fig. 3). Nitroprusside infusion did not affect EhI (Fig. 3), but induced a significant fall in Hct (—6 2%). In contrast with ERPFhIP, RBFhiP did not change significantly (+11 5%).
60 40 *
20
Indomethacin. In response to indomethacin (1 mg kg') MAP
0
and RVR remained unchanged (—6 4 and +2 5%). Acute inhibition of prostaglandin synthesis, however, resulted in a fall,
—20
non-significantly, in RBFprobe and, significantly, in ERPFhIP (Fig. 3). EhIP did not change (Fig. 3), as was true for Hct (—1 3%).
B
60 C U)
20
0
RBFhIP fell (—8
2%). Angiotensin II. Infusion of 0.51 0.08 xg kg1 hr1 angiotensin II resulted in a rise of MAP (+14 3%), as well as
I.H
40
0.-a
RVR (+56 6%). Both RBFpr,)hC and ERPFhIP fell, but EhIP rose (Fig. 3). Hct was not affected by infusion of angiotensin II (+0
2%). RBFhIP, however, fell (—24 2%). *
—20
Relation between changes in RBFP,Oh and changes in plasma hippurate clearance
C
10 *
5.
II
0—5 -
—10-
Linear regression analysis showed that drug-induced changes in ERPFhP correlated well with changes in RBFPtOhe (r = 0.902, P < 0.01; Fig. 4A). The slope, however, was significantly different from the line of identity (0.562, 95% CI 0.463 to 0.661), indicating that
changes, either a rise or fall, in ERPFhP were less pronounced in comparison with changes in RBFPrObC. Stepwise correction of changes in ERPFhIP for Hct (Fig. 4B), for EhIP (Fig. 4C), or for
both parameters (Fig. 4D) improved the relation such that, *
ACEi DOP NE
NIP
Fig. 3. Placebo corrected percentile changes (mean
IND Ang II SEM)
in (A) left renal
blood flow measured by flow probe (RB! ,h), (B) plasma hippurate
clearance (ERPFhiP), and (C) renal hzpurate extraction ratio (E,,1) induced by enalaprilate (ACEi), dopamine (DOP), norepinephrine (NE), nitropmsside (NIP), indomethacin (IND), and angiotensin II (Ang II) (*P < 0.05 vs. placebo).
2, RBFprObC and ERPFhIP also did not change significantly (—1 1 and —6 2%). EhI, however, fell significantly during the course
indeed, changes in RBFhiP were nearly identical with changes in RBFPrObC (slope 0.817, 95% CI 0.731 to 0.903; r = 0.961, P < 0.01). Interestingly, changes in EhIP correlated inversely with changes in RBFPTOhe (r = —0.789, P < 0.01), whereas changes in Hct did not (r = 0.204, P > 0.05). Multiple regression analysis
showed that changes in EhP did not independently affect the
relation between drug-induced changes in ERPFhIP and changes in RBFPr0bc. This indicates that in estimating changes in RBF by means of the hippurate clearance technique, EhIP can be substi-
tuted by a constant regression coefficient of ERPFhIP. This observation contrasts with the finding that changes in Hct did independently contribute to the relation between changes in
ERPFhIP and changes in RBFProbe (P < 0.01). This implies that in of the experiments (—7 1%). Hct was not affected by placebo using changes in hippurate clearance as a measure of changes in 2%), as was true for RBFhIP (—0 infusion (—3 2%). RBF, a correction for changes in Hct needs to be performed. Drug-induced changes as discussed in the following paragraphs Thus, the present data indicate that for practical purposes changes are corrected for these placebo effects. in RBF can be calculated (RBFCaIC) without knowledge of EhIP, Enalaprilate. In response to ACEi (10 mg) both MAP and RVR with the following formula: fell (—14 1 and —40 6%). RBFPrObe and ERPFhP both rose, whereas EhIP fell (Fig. 3). ACEi did not affect Hct (+1 3%), butRBF(% of baseline) = 1.529 X zERPF(% of baseline) + 1.296 induced a rise in RBFhP (+35 9%). x Hct(% of baseline), as shown in Figure 5 (r = 0.939, P < 0.01). Dopamine. In response to infusion of 0.29 0.03 mg kg
hr dopamine, MAP remained unchanged (—2 2%), whereas
Discussion RVR fell (—27 4%). Both RBFprohe and ERPFhIP rose and EhIP We compared drug-induced changes in renal blood flow meafell (Fig. 3). Furthermore, dopamine infusion induced a significant rise in Hct and in RBFhIP (+2 1 and +28 6%). sured by renal blood flow probe (RBFPr0bc) and by plasma
Norepinephrine. Infusion of 0.02 mg kg1 - hr 1 resulted in a significant rise of MAP (+17 2%), whereas, intriguingly, RVR remained unchanged (—1 2%). RBFPrObC rose, while ERPFhIP and EhI remained stable (Fig. 3). Hct rose (+6 1%), whereas RBFhIP rose non-significantly (+6 4%).
hippurate clearance (ERPFhIP). Changes in RBFPrOhe and ERPFhIP correlated well (r = 0.902, P < 0.01), although changes in ERPFhIP were less pronounced in comparison with changes in RBFPrOhe. Correction of changes in ERPFhIP for renal hippurate extraction (EhI) and hematocrit (Hct) improved the relation with
1621
Visscher et al. Hippurate clearance and renal blood flow
A
B
C
D
80
40
•0 a) a) U) Ca
-40 a)
0C
80 Fig. 4. Correlations between dmg-induced changes in plasma hippurate clearance and changes in left renal blood flow measured by renal a/Ic!)) probe (RBFP,,b). (A) Changes in plasma hippurate clearance without Correction (r = 0.902). (B) Changes in plasma hippurate clearance corrected for Hct (r = 0.932). (C) Changes in plasma hippurate clearance corrected for Eh (r = 0.942). (D) Changes in plasma hippurate clearance corrected for both Hct and EhIP (r 0.961). Changes in response
40
0
—40
to enalaprilate (0), dopamine (•),
—40
0
40
80
—40
0
i RBFpo (% of baseline)
40
80
norepinephrine (A), nitroprusside (A),
indomethacin (0), and angiotensin II (•). The solid line indicates the line of identity, the dotted line indicates the regression line.
changes in RBFprobe. However, since changes in EhIP were inversely related to changes in RBFPrObC, only changes in Hct
without knowledge of EhIP, drug-induced changes in plasma hippurate clearance corrected for changes in Hct are a good affected the relation between changes in ERPFhP and RBFPrObe means to establish changes in renal blood flow. In doing so it independently. This indicates that for practical purposes drug- should be kept in mind that the presented algorithm is based upon induced changes in ERPFhIP, corrected for Hct, are a good means the existence of a negative relation between changes in renal to establish changes in renal blood flow. An objection that might be raised against our study is the fact that RBFprobe was established for one kidney only. Since druginduced changes in RBF are very likely to be the same in both kidneys, relative changes in one kidney RBFPrObe can well be
blood flow and changes in
compared with changes in plasma hippurate clearance. The
during ACEi and dopamine, whereas EhP rose in response to
present data show that a good relation exists between the druginduced relative changes in ERPFhIP and relative changes in one kidney RBFprobe. Drug-induced changes in ERPFhIP, however, were less pronounced than those in RBFprObe. This is a phenom-
angiotensin II. The findings on the changes induced by ACEi and
In renal disease, such as renovas-
cular hypertension [3, 12], this relation may be disturbed and therefore the computation is no longer valid. How can we explain the observed changes in renal hippurate extraction? In the present study EhP fell during placebo as well as
angiotensin II are comparable with previous findings on renal extraction of paraaminohippurate and orthoiodohippurate [2, 3], whereas the dopamine induced fall in EhI to our knowledge has enon which, in agreement with others [6], can be largely overcome never been reported. Several factors have been proposed to affect by correcting ERPFhP for Eh and Hct. However, the present renal extraction of hippurate or its derivatives [13]. First, a data also indicate that EhIP does not independently affect the pronounced fall in glomerular filtration rate (GFR) might induce relation between changes in ERPFhIP and changes in RBFprobe a fall in EhIP [14]. Second, EhIP might fall due to inhibition of the and can thus be substituted by a simple correction factor. This tubular transport mechanism by competition between drug and observation has an important clinical impact: the determination of hippurate for the organic anion transport mechanism [15—171. EhIP requires renal vein catheterization. Hct measurements, on Third, renal tracer extraction may be changed by alterations in the other hand, are relatively simple to perform. Since changes in intrarenal blood flow distribution [2, 5, 18, 19]. Fourth, due to this parameter independently affect the relation between changes changes in plasma transit time, EhP might change inversely with in ERPFhP and changes in RBFPrObe, changes in Hct must be changes in RBF [4, 5, 201. Lastly, it has been suggested that the established to obtain adequate results. Thus, our data show that rate of renal tracer secretion is enhanced by an increase in sodium
1622
Visscher et al: Hippurate clearance and renal blood flow
Selig showed, as we did, in conscious dogs using Doppler flow probes, that intravenous infusion of norepinephrine induces a rise
80
in RBF [32]. They attributed this RBF increase to a reflex mediated vasodilation, a mechanism which is likely to be inhibited during anaesthesia [33]. Another interesting example of anaesthesia/drug-effect interaction is the study of Swain et al [34] in dogs, who showed that during anaesthesia indomethacin induced a fall in RBF (probes), whereas during consciousness this fall was not
a)
C
a) U)
40
CO
0 U-
0
—40
—40
0
40
80
A RBFprobe (% of baseline)
Fig. 5. Correlation between drug-induced calculated changes in RBF (RBF1), based upon changes in ERPF,J, and changes in Hct, and changes in left renal blood flow measured by renal artely probe (RBF,.0b,.). Changes in response to enalaprilate (0), dopamine (•), norepinephrine (A), nitroprusside (h), indomethacin (Eli), and angiotensin II (•). The solid line indicates the line of identity, the dotted line indicates the regression line.
reabsorption rate [21, 22]. The observed fall in EhiP during placebo cannot be attributed to changes in renal function since
significant. In the present study, indomethacin indeed induced a non-significant fall in RBFPrObC. Thus, the present data once more stress that one should be careful when interpreting results from experiments during anaesthesia. In conclusion, the present data indicate that changes in plasma hippurate clearance are a good means to establish drug-induced relative changes in renal blood flow. A simple algorithm which takes changes in hematocrit into account suffices for the conversion of hippurate clearance data to changes in renal blood flow. If the algorithm is used, knowledge of the actual renal hippurate extraction is not needed. Since drug effects on plasma hippurate clearance, renal hippurate extraction, and hematocrit reported are comparable in dog and humans, extrapolation of the algorithm to humans appears valid. Therefore, renal vein catheterization to obtain renal hippurate extraction is unnecessary in this setting. Acknowledgments This study was made possible by grant (C92.126l) of the Dutch Kidney Foundation. We are indebted to Mr. J.M. Elstrodt for technical assistance, and to Mr. A. Nijmeijer for assistance in training the dogs.
Reprint requests to R.M. Huisman, M.D., Department of Medicine, RBF as well as RVR remained stable. Tubular hippurate secreof Nephrology, University Hospital, P.O. Box 30.001, 9700 RB tion, however, may have been inhibited by the glucose infusion Division Groningen, The Netherlands. either directly by competition [7] or indirectly by a fall in sodium reabsorption rate via extracellular volume expansion [22]. AnReferences other possibility is redistribution of renal blood flow in favor of the medulla due to renal nerve stimulation [23]. Both explanations are speculative. Changes in cortical transit time as a cause of 1. SMITH HW: The Kidney: Structure and Function in Health and Disease.
changes in EhIP appears a good overall explanation for the drug-induced changes in EhI, since in the present data a clear negative relation exists between changes in EhP and changes in RBFPrObC. Based upon the present results, however, it cannot be concluded what, if any, overall explanation is the correct one.
The hemodynamic effects of the studied vasoactive drugs compare well with previous data both in dog and man despite considerable differences in study design. As previously shown, ACEi induced a fall in systemic blood pressure (MAP) and renal vascular resistance (RVR), and a rise in RBF established either by
flow probe or by hippurate clearance [3, 24, 25]. The effects of angiotensin II were exactly the opposite [26]. Dopamine did not
affect MAP, but induced a fall in RVR and a rise in RBF independent of the technique used [27, 28]. In response to nitroprusside both MAP and RVR fell, whereas RBF, measured either way, did not change significantly [29, 301. The observation
that intravenously infused norepinephrine induced a rise in RBFProhe without a change in RVR, however, contrasts with the general opinion that norepinephrine induces a fall in RBF with an increase in RVR [31]. The latter conclusion is based on studies performed during anaesthesia and/or in which RBF was measured by means of clearance techniques. Indeed, Anderson, Korner and
New York, Oxford University Press, 1951 2. CARRIERE S, FRIBORO J: Intrarenal blood flow and PAH extraction during angiotensin infusion. Am J Physiol 217:1708—1715, 1969 3. WENTING GJ, TAN-TJONG HL, DERKX FHM, DE BRUYN JHB, MAN
IN 'T VELD AA, SCHALEKAMP MADH: Split renal function after captopril in unilateral renal artery stenosis. Br Med J 288:886—890, 1984 4. VELASQUEZ MT, NOTARGIACOMO AV, COHN JN: Influence of cortical
plasma transit-time on p-aminohippurate extraction during induced renal vasodilatation in anaesthetized dogs. C/in Sci 43:401—411, 1972 5. AUKLAND K, LØYNING EW: Intrarenal blood flow and para-aminohip-
purate (PAll) extraction. Acta Physiol Scand 79:95—108, 1970 6. LEVINSKY NG, LIEBERTHAL W: Clearance techniques, in Renal Physiology, Handbook of Physiology, section 8, edited by WINDHAGER EE,
New York, Oxford University PreSS, 1992, p 236 7. LOTE CJ, MCVICAR AJ, YARDLEY CP: Renal extraction and clearance
of p-aminohippurate during saline and dextrose infusion in the rat. J Physiol 363:303—313, 1985
8. ZWEENS J, SCHIPHOF P: Permanent catheterization of aorta and pulmonary artery in the dog. Pfldgers Arch 362:201—202, 1976 9. Ross J JR COVELL JW, FELD OK, SCHMID-SCI-IOENBEIN G: Cardio-
vascular system, in Best and Taylor's Physiological Basis of Medical Practice, edited by WEsT JB, Baltimore, Williams and Wilkins, 1990, pp 245—246
10. JONKER GJ, Dc ZEEUw D, HUISMAN RM, VAN DER HEM GK: Pharmacological zero for electromagnetic renal blood flow measurement. Pflugers Arch 403:220—221, 1985
Visscher et a!: Hippurate clearance and renal blood flow
1623
11. DONKER AJM, VAN DER HEM GK, SLUITER WJ, BEEKHUIS H: A
the intrarenal distribution of blood during the natriuresis accompa-
radioisotope method for simultaneous determination of the glomerular filtration rate and the effective renal plasma flow. Neth J Med
flying saline loading in the dog. J Clin Invest 44:929—941, 1965 23. POMERANZ BH, BIRTCH AG, BARGER AC: Neural control of intrarenat blood flow. Am J Physiol 215:1067—1081, 1968 24. HOLLENBERG NK, MEGGS LG, WILLIAMS GH, KATZ J, GARNIC JD, HARRINGTON DP: Sodium intake and renal responses to captopril in
20:97—103, 1977
12. VI5scHER CA, DE ZEEUW D, DE JUNG PE, PIERS DA, BEEKHUIS H,
HuISMAN RM. ACEI-induced changes in hippurate handling in renovascular hypertension. (abstract) Kidney mt 46:1472, 1994 13. MØLLER JV, SHEIKH MI: Renal organic anion transport system: pharmacological, physiological, and biochemical aspects. Pharmacol Rev 34:315—358, 1983 14. HABERLE DA, RUHLAND G, LAUSSER A, MOORE L, NEISS A: Influence
of glomerular filtration rate on renal PAH secretion rate in the rat kidney. Pflugers Arch 375:131—139, 1978 15. SINGHVI SM, DUCHIN KL, WILLARD DA, MCKINSTRY DN, MIGDALOF
BH: Renal handling of captopril: Effect of probenecid. Clin Pharmacol Ther 32:182—189, 1982
16. ULLRICH KJ, RUMRICH C, DAVID C, FRITZSCH G: Bisubstrates: Substances that interact with renal contraluminal organic anion and
cation transport systems. I. Amines, piperidines, piperazines, azepines, pyridines, quinolines, imidazoles, thiazoles, guanidines and hydrazines. Pflugers Arch 425:280—299, 1993 17. ULLRICH KJ, RUMRICH C, DAVID C, FRITZSCH G: Bisubstrates: Substances that interact with both, renal contraluminal organic anion and cation transport systems. II. Zwitterionic substrates: Dipeptides, cephalosporins, quinolone-carboxylate gyrase inhibitors and phospha-
mide thiazine carboxylates; nonionizable substrates: Steroid hor-
mones and cyclophosphamides. Pflugers Arch 425:300-312, 1993 18. PILKINGTON LA, BINDER R, DE HAAs JCM, Prrrs RF: Intrarenal distribution of blood flow. Am J Physiol 208:1107—1113, 1965
19. REUBI F: Objections a Ia théorie de Ia separation intrarbnale des hématies et du plasma (Pappenheimer). Helv Med Acta 4:516—523, 1958
20. CARRIERE 5, FRIBORG J, GUAY JP: Vasodilators, intrarenal blood flow, and natriuresis in the dog.AmJPhysiol 221:92—98, 1971 21. VOGEL G, KROGER W: Die Bedeutung des Transportes, der Konzen-
tration und der Darbietungsrichtung von Na für den tubularen Glucose- und PAH-transport. Plugers Arch 288:342—358, 1966 22. EARLY LE, FRIEDLER RM: Changes in renal blood flow and possibly
normal man and in essential hypertension. Kidney mt 20:240—245, 1981
25. ZIMMERMAN BG, GOERING JL, RAICH PC: Resistance to blockade by
saralasin of effect of ACE inhibitors in conscious sodium-restricted dogs. Am J Physiol 255:F944—F952, 1988 26. MITCHELL KD, NAVAR LG: Interactive effects of angiotensin II on renal haemodynamics and tubular reabsorptive function. Kidney mt 38(Suppl 30):69—73, 1990 27. MCDONALD RH JR, GOLDBERG LI, MCNAY JL: Effects of dopamine
in man; augmentation of sodium excretion, glomerular filtration rate and renal plasma flow. J Clin Invest 43:1116—1124, 1964 28. MCNAY JL, MCDONALD RH JR, GOLDBERG LI: Direct vasodilation produced by dopamine in the dog. Circ Res 16:510—517, 1965 29. NIELSEN CB, EISKJIER H, PEDERSEN EB: Enhanced renal production
of cyclic GMP and reduced free water clearance during sodium nitroprusside infusion in healthy man. Eur J Clin Invest 23:375—381, 1993 30. PAGANI M, VATNER SF, BRAUNWALD E: Hemodynamic effects of
intravenous sodium nitroprusside in the conscious dog. Circulation 57:144—151, 1978
31. SCHRIER RW: Effects of adrenergic nervous system and catecholamines on systemic and renal hemodynamics, sodium and water excretion and renin secretion. Kidney mt 6:291—306, 1974 32. ANDERSON WP, KORNER PT, SELIG SE: Mechanism involved in the renal responses to intravenous and renal artery infusions of noradrenaline in conscious dogs. J Physiol 321:21—30, 1981 33. VATNER SF, BRAUNWALD E: Cardiovascular control mechanisms in the conscious state. N Engi J Med 293:970—976, 1975 34. SWAIN JA, HEYNDRICKX GR, BOETFCHER DH, VATNER SF: Prosta-
glandin control of renal circulation in the unanesthetized dog and baboon. Am J Physiol 229:826—830, 1975
Drug-induced changes in renal hippurate clearance as a measure
of renal blood flow C. ANTOINETFE VISSCHER, DICK DE ZEEUW, PAUL E. DE JONG, WIM J. SLUITER,
and ROEL M. HUISMAN Groningen Institute for Dnig Studies1 (GIDS), Department of Medicine, Divisions of Nephrology and Endocrinology, University Hospital, Groningen, The Netherlands
Drug-induced changes in renal hippurate clearance as a measure of renal blood flow. We studied the accuracy of the plasma 1311-hippurate clearance technique to monitor drug-induced changes in renal blood flow
(RBF) by comparing it to a flow probe technique in six conscious, chronically instrumented dogs. Placebo caused no change in RBF, either established by hippurate clearance (ERPF5) or by renal blood flow probe (RBFproh,,). Enalaprilate induced a rise in ERPFhIP and RBFPrObC (+26 5 and 44 12%), as did dopamine (+16 4 and +33 5%). Intravenous
infusion of norepinephrine induced a rise in ERPFhIP (+2 6%, NS) and
in RBFprob,, (+18 3%), as did nitroprusside (+14 4% and +13 6%,
may still be accurately measured using renal hippurate clearance without knowing Eh1P, presuming that the intervention does not change EhIP. Yet, vasoactive substances are reported to affect EhIP [2—41. In addition, changes in hematocrit may occur affecting the
relation between RBF and hippurate (plasma) clearance. Thus, (drug)-induced changes in renal hippurate clearance are commonly considered to be an imprecise indicator even of changes in RBF [5—7]. To our knowledge, however, no studies have directly
NS). Indomethacin induced a fall in ERPFhP (—8 2%) and in RBFprObc
compared the clearance technique with the gold standard, the
angiotensin II (—5 2, —7 1, and +5 2%, respectively). Hematocrit (Hct) was affected by dopamine, norepinephrine, and nitroprusside (+2 1, +6 1, and —6 2%, respectively). Drug-induced changes in ERPF5Ip
renal artery blood flow probe, in measuring drug-induced changes in RBF and evaluated the impact of changes in EhIP or hematocrit on the measurement. In the present study we therefore compared changes in directly measured RBF with changes in plasma hippurate clearance with and without correction for EhIP and/or hematocrit in chronically
(—7 3%, NS), as did angiotensin II (—19 1 and —26 3%). Renal hippurate extraction (EhI) was affected by enalaprilate, dopamine, and
correlated well with changes in RBFPr,b (r = 0.902, P < 0.01). Changes in E1 did not independently affect this relation, whereas changes in Hct did: RBF(% of baseline) = 1.529 X IIERPFSiP(% of baseline) + 1.296 X zHct(% of baseline). These data indicate that drug-induced changes in plasma hippurate clearance can, even when changes in renal hippurate extraction are unknown, be used as a reliable indicator of changes in renal blood flow if changes in hematocrit are taken into account.
instrumented conscious dogs. These changes were induced by different drugs and hormones which are of clinical importance (enalaprilate, dopamine, norepinephrine, nitroprusside, indomethacin, and angiotensin II).
The renal clearance of exogenously infused hippurate has
Methods Six male mongrel dogs (25 to 33 kg body wt) were trained to become accustomed to the experimental situation and were kept
hippurate. Some hippurate, however, appears to escape renal
Instrumentation was performed in two steps under general
frequently been used since the fifties [1] as a means to establish on a salt restricted diet of less than I g of salt per day (H/D renal blood flow (RBF) both in the clinical and the laboratory Prescription Diet, Hill's Pet Products, Topeka, KS, USA). setting. This application is based on the presumption that plasma Instrumentation passing the kidney is nearly completely and constantly cleared of
anaesthesia (induction with thiopental, maintenance anaesthesia [1]. Thus, renal hippurate clearance can only be used as an with nitrous oxide and isoflurane). First, the left renal artery was absolute measure of RBF when corrected for EhIP. In practice this exposed retroperitoneally through a flank incision and dissected is not often done since EhP determination requires an invasive from surrounding tissue. Subsequently, a transit-time flow probe catheterization of the renal vein. Despite this, acute or even (Transonic Systems, Ithaca, NY, USA) of appropriate size was chronic changes in RBF induced by drug or other interventions placed around the renal artery. Furthermore, Tygon catheters (Norton, Akron, OH, USA; o.d. 1.8 mm, i.d. 1.0 mm) were placed in the descending aorta and the right atrium through the omocervical artery and vein, respectively [8]. Second, after a period of about two weeks recovery, the left renal vein was exposed through GIDS is part of the research school, Groningen Utrecht Institute for a midline incision and the spermatic vein was ligated to prevent Drug Exploration (GUIDE) sampling of non-renal blood. A heparin-coated Tygon catheter Received for publication February 6, 1995 and in revised form June 2, 1995 was placed in the renal vein through and fixed onto the inferior Accepted for publication June 2, 1995 caval vein. All leads were tunnelled subcutaneously to a location high on the back of the animal, and covered by a jacket. The © 1995 by the International Society of Nephrology clearance, so that renal hippurate extraction (EhIP) is never 100%
1617
1618
Visscher et al: Hippurate clearance and renal blood flow
collected by bladder catheterization. After a priming dose, 1311
E
300
hippurate was infused at a rate of 12 ml hr1 according to the method described by Donker et al [11]. After a stabilization period of two hours, six half-hour renal clearance periods were
200
performed: three baseline measurements followed by three drug or placebo measurements (starting 1 hour from administration). To measure renal hippurate extraction, arterial and renal venous blood samples were always drawn in the same order within three minutes of each other. Blood samples were immediately centrifuged at 4°C (480 g) and plasma was removed immediately to prevent diffusion of hippurate from the red cells into plasma.
100
Before the baseline and after the experimental period arterial
pa)0
0.
U-
blood samples were drawn to establish hematocrit (Hct). During
the baseline as well as during the experimental period mean arterial pressure (MAP: Disposable Pressure Transducer, Cobe® Laboratories) and left renal blood flow (RBFprobe: renal artery probe) were continuously registered.
0
0
100
200
300
RBFVO mi/mm
Fig. 1. Convlation between left renal blood flow measured by flow probe (RBFP,,h) and established by renal venous outflow (RBF,,,) in two dogs (• and 0). The solid line indicates the line of identity.
Data analysis Renal vascular resistance was calculated as MAP divided by RBFprobe.
The extraction ratio of hippurate was calculated as the difference between arterial (AhP) and renal venous 1311-hippurate concentration (Vh,) divided by Ah)P:
Eh1 =
(A
—
VP)/AhP
position of the renal vein catheter was verified before each Renal clearance of '311-hippurate was calculated according the experiment by oxygen saturation measurement [9]. All catheters following formula:
were kept patent by refilling them daily with a fresh heparin
solution (2000 IE/ml). Flow probes were precalibrated by the producer. In two dogs under general anaesthesia calibration was verified in vivo after the final experiment. This, by comparing flow probe values with renal venous outflow measurements, Changes in renal blood flow were induced by bleeding: One flow probe constantly overestimated (22 7%) whereas the other flow probe slightly underestimated (—9
ERPFh1(U) =
unneblP
X urine flow rate/A
Plasma clearance of hippurate was calculated as follows:
ERPF(I) = infusate X infusion rate/A Renal blood flow (RBFhIP) was subsequently calculated as:
5%) renal blood flow. Furthermore, calibration curves were RBFhP = ERPFhP/{EIJP x (1 — Hct)] perfectly linear (r = 0.98 and >0.99; Fig. 1). By intra-arterially injected angiotensin II (5 to 10 j.tg) it was verified in vivo that no The means of the three clearance periods before and after drug or placebo administration were used as the respective baseline drifting of zero level occurred [10]. SEM. experimental values. Values are expressed as mean Experimental design
Statistical analysis was performed with the commercially available
After completion of the instrumentation the animal was allowed to recover for a least one week. The acute effects of placebo, enalaprilate (ACEi), dopamine (DOP), norepinephrine (NE), nitroprusside (NIP), indomethacin (IND), and angiotensin II (Ang II) were studied on subsequent experimental days in a randomized order. Placebo, ACEi (10 mg enalaprilate i.v.) and IND (1 mg kg' Indocid® P.D.A. i.v.; a gift of MSD, Haarlem, The Netherlands) were given in a single dosage, whereas DOP, NE, NIP, and Ang II were continuously infused (15 ml hr1 i.v.)
software program Instat (GraphPad Software, San Diego, CA, USA). Differences between baseline and experimental values during placebo were tested by the Student's paired t-test. Druginduced changes were expressed as percentage change from baseline, and subsequently tested by comparing them with the placebo values by the paired t-test. The relation between changes
at a dose titrated to a 10 to 30% change in mean arterial pressure and renal blood flow. Drugs were dissolved in 5% glucose. To
Results
prevent drug interaction or carry-over, the drugs were studied during at least one week intervals. Before the experimental day dogs had free access to water, but they were deprived of food for at least eight hours. During the experiment the trained dog was conscious and hanging quietly in a hammock. To provide adequate urine production 5% glucose was infused continuously at a rate of 250 ml hr'. Urine was
in ERPFhIP, EhiP, and Hct and changes in RBFPrObC was tested by
multiple variable regression analysis. Differences and correlations were considered significant at a level of 5%.
Baseline values
Table 1 shows the baseline values of all parameters established during the experimental days with placebo (N = 6), ACEi (N = 5), dopamine (N = 5), norepinephrine (N = 5), nitroprusside (N = 6), indomethacin (N = 6), and angiotensin II (N = 6). No significant differences were observed, although baseline values
varied slightly between the experimental days. Furthermore,
1619
J/isscher et a!: Hippurate clearance and renal blood flow
Table 1. Baseline values of mean arterial pressure (MAP), renal vascular resistance (RVR), left renal blood flow (RBFprObe), plasma [ERPFhIP (I)] and renal hippurate clearance [ERPFhP (U)I, renal hippurate extraction (Eh,P), haematocrit (Hct), and renal blood flow as established by plasma hippurate clearance (RBFhIP)
MAP mm Hg Plac
ACEi DOP NE NIP IND
Ang II
113 109 113 109 109 111 109
0.505 0.415
Hct
mi/mm
246 8
0.460 0.023 0.449 0.042 0.480 0.046
5 6 5 6 6 5 4
ERPFhIP (U)
ERPFhIP (I)
RBFprOb
RVR mm Hg/mm/mi
0.075 0.048
0.446 0.051 0.416 0.044
277
306 26
305 27 294 34 291 20 294 33 322 35 325 24 322 32
248 16 241 21 225 24 276 29 265 30 30
291
0.77
0.02
297 28 297 33 316 36 341 27 314 34
0.77 0.78 0.75 0.74
0.02 0.02 0.03 0.02
mI/mm
39 3 42 1 40 3 39 1 40 2 38 2 40 2
0.77 0.02 0.75 0.02
34
RBFhP
%
EhI
618 43 655 59 639 27 646 80 693 67 701 46 726 62
Values shown during placebo (Plac), enalaprilate (ACEi), dopamine (DOP), norepinephrine (NE), nitroprusside (NIP), indomethacin (IND), and angiotensin II (Ang II) experiments.
Placebo
IND
NIP
NE
DOP
ACEI
Ang II
—
400 E .0 0 0.
U-
200 -
400 E
0.
0
-
U-
200 -
E *#
'I
0.80 0.
*
w 0.60 -
+
—
+
—
+
+
—
—
+
—
+
—
+
Fig. 2. Individual values of left renal blood flow measured by flow probe (RBFP,.c,b), plasma hippurate clearance (ERPFh,I,), and renal hippurate e.rtraction ratio (E1,q,), before and after administration of placebo, enalaprilate (ACEi), dopamine (DOP), norepmnephrine (NE), nitroprusside (NIP), indomethacin
(IND), or angiotensin II (Ang II) (*P < 0.05 vs. baseline, #P < 0.05 vs. placebo).
baseline values of ERPFhIP(I) and ERPFhiP(U) were comparable.
This indicates that at steady state the appearance rate of hippurate in the plasma equals the appearance rate in the urine, and
Drug-induced changes
Figure 2 shows the individual values of RBFPrOBe, ERPFhIP, and
that ERPFhP(I) can be substituted by ERPFhIP(U). Since EhIP on the different experimental days, whereas in Figure 3 the ERPFhP(I) is not subject to errors introduced by incomplete average placebo corrected changes are shown. urine collection or dead-space, this entity was used to establish drug-induced changes in renal hippurate clearance.
Placebo. Placebo infusion did not induce significant changes in MAP and RVR (+2 1 and +3 2%). As can be seen in Figure
1620
Visscher et a!: Hippurate clearance and renal blood flow
A
Nitroprusside. Nitroprusside was given at a mean dosage of 0.29 0.03 mg kg1 - hr'. This resulted in a fall in MAP as well as RVR (—23 2 and —31 6%). RBFprobe was not significantly affected, whereas ERPFhIP rose significantly (Fig. 3). Nitroprusside infusion did not affect EhI (Fig. 3), but induced a significant fall in Hct (—6 2%). In contrast with ERPFhIP, RBFhiP did not change significantly (+11 5%).
60 40 *
20
Indomethacin. In response to indomethacin (1 mg kg') MAP
0
and RVR remained unchanged (—6 4 and +2 5%). Acute inhibition of prostaglandin synthesis, however, resulted in a fall,
—20
non-significantly, in RBFprobe and, significantly, in ERPFhIP (Fig. 3). EhIP did not change (Fig. 3), as was true for Hct (—1 3%).
B
60 C U)
20
0
RBFhIP fell (—8
2%). Angiotensin II. Infusion of 0.51 0.08 xg kg1 hr1 angiotensin II resulted in a rise of MAP (+14 3%), as well as
I.H
40
0.-a
RVR (+56 6%). Both RBFpr,)hC and ERPFhIP fell, but EhIP rose (Fig. 3). Hct was not affected by infusion of angiotensin II (+0
2%). RBFhIP, however, fell (—24 2%). *
—20
Relation between changes in RBFP,Oh and changes in plasma hippurate clearance
C
10 *
5.
II
0—5 -
—10-
Linear regression analysis showed that drug-induced changes in ERPFhP correlated well with changes in RBFPtOhe (r = 0.902, P < 0.01; Fig. 4A). The slope, however, was significantly different from the line of identity (0.562, 95% CI 0.463 to 0.661), indicating that
changes, either a rise or fall, in ERPFhP were less pronounced in comparison with changes in RBFPrObC. Stepwise correction of changes in ERPFhIP for Hct (Fig. 4B), for EhIP (Fig. 4C), or for
both parameters (Fig. 4D) improved the relation such that, *
ACEi DOP NE
NIP
Fig. 3. Placebo corrected percentile changes (mean
IND Ang II SEM)
in (A) left renal
blood flow measured by flow probe (RB! ,h), (B) plasma hippurate
clearance (ERPFhiP), and (C) renal hzpurate extraction ratio (E,,1) induced by enalaprilate (ACEi), dopamine (DOP), norepinephrine (NE), nitropmsside (NIP), indomethacin (IND), and angiotensin II (Ang II) (*P < 0.05 vs. placebo).
2, RBFprObC and ERPFhIP also did not change significantly (—1 1 and —6 2%). EhI, however, fell significantly during the course
indeed, changes in RBFhiP were nearly identical with changes in RBFPrObC (slope 0.817, 95% CI 0.731 to 0.903; r = 0.961, P < 0.01). Interestingly, changes in EhIP correlated inversely with changes in RBFPTOhe (r = —0.789, P < 0.01), whereas changes in Hct did not (r = 0.204, P > 0.05). Multiple regression analysis
showed that changes in EhP did not independently affect the
relation between drug-induced changes in ERPFhIP and changes in RBFPr0bc. This indicates that in estimating changes in RBF by means of the hippurate clearance technique, EhIP can be substi-
tuted by a constant regression coefficient of ERPFhIP. This observation contrasts with the finding that changes in Hct did independently contribute to the relation between changes in
ERPFhIP and changes in RBFProbe (P < 0.01). This implies that in of the experiments (—7 1%). Hct was not affected by placebo using changes in hippurate clearance as a measure of changes in 2%), as was true for RBFhIP (—0 infusion (—3 2%). RBF, a correction for changes in Hct needs to be performed. Drug-induced changes as discussed in the following paragraphs Thus, the present data indicate that for practical purposes changes are corrected for these placebo effects. in RBF can be calculated (RBFCaIC) without knowledge of EhIP, Enalaprilate. In response to ACEi (10 mg) both MAP and RVR with the following formula: fell (—14 1 and —40 6%). RBFPrObe and ERPFhP both rose, whereas EhIP fell (Fig. 3). ACEi did not affect Hct (+1 3%), butRBF(% of baseline) = 1.529 X zERPF(% of baseline) + 1.296 induced a rise in RBFhP (+35 9%). x Hct(% of baseline), as shown in Figure 5 (r = 0.939, P < 0.01). Dopamine. In response to infusion of 0.29 0.03 mg kg
hr dopamine, MAP remained unchanged (—2 2%), whereas
Discussion RVR fell (—27 4%). Both RBFprohe and ERPFhIP rose and EhIP We compared drug-induced changes in renal blood flow meafell (Fig. 3). Furthermore, dopamine infusion induced a significant rise in Hct and in RBFhIP (+2 1 and +28 6%). sured by renal blood flow probe (RBFPr0bc) and by plasma
Norepinephrine. Infusion of 0.02 mg kg1 - hr 1 resulted in a significant rise of MAP (+17 2%), whereas, intriguingly, RVR remained unchanged (—1 2%). RBFPrObC rose, while ERPFhIP and EhI remained stable (Fig. 3). Hct rose (+6 1%), whereas RBFhIP rose non-significantly (+6 4%).
hippurate clearance (ERPFhIP). Changes in RBFPrOhe and ERPFhIP correlated well (r = 0.902, P < 0.01), although changes in ERPFhIP were less pronounced in comparison with changes in RBFPrOhe. Correction of changes in ERPFhIP for renal hippurate extraction (EhI) and hematocrit (Hct) improved the relation with
1621
Visscher et al. Hippurate clearance and renal blood flow
A
B
C
D
80
40
•0 a) a) U) Ca
-40 a)
0C
80 Fig. 4. Correlations between dmg-induced changes in plasma hippurate clearance and changes in left renal blood flow measured by renal a/Ic!)) probe (RBFP,,b). (A) Changes in plasma hippurate clearance without Correction (r = 0.902). (B) Changes in plasma hippurate clearance corrected for Hct (r = 0.932). (C) Changes in plasma hippurate clearance corrected for Eh (r = 0.942). (D) Changes in plasma hippurate clearance corrected for both Hct and EhIP (r 0.961). Changes in response
40
0
—40
to enalaprilate (0), dopamine (•),
—40
0
40
80
—40
0
i RBFpo (% of baseline)
40
80
norepinephrine (A), nitroprusside (A),
indomethacin (0), and angiotensin II (•). The solid line indicates the line of identity, the dotted line indicates the regression line.
changes in RBFprobe. However, since changes in EhIP were inversely related to changes in RBFPrObC, only changes in Hct
without knowledge of EhIP, drug-induced changes in plasma hippurate clearance corrected for changes in Hct are a good affected the relation between changes in ERPFhP and RBFPrObe means to establish changes in renal blood flow. In doing so it independently. This indicates that for practical purposes drug- should be kept in mind that the presented algorithm is based upon induced changes in ERPFhIP, corrected for Hct, are a good means the existence of a negative relation between changes in renal to establish changes in renal blood flow. An objection that might be raised against our study is the fact that RBFprobe was established for one kidney only. Since druginduced changes in RBF are very likely to be the same in both kidneys, relative changes in one kidney RBFPrObe can well be
blood flow and changes in
compared with changes in plasma hippurate clearance. The
during ACEi and dopamine, whereas EhP rose in response to
present data show that a good relation exists between the druginduced relative changes in ERPFhIP and relative changes in one kidney RBFprobe. Drug-induced changes in ERPFhIP, however, were less pronounced than those in RBFprObe. This is a phenom-
angiotensin II. The findings on the changes induced by ACEi and
In renal disease, such as renovas-
cular hypertension [3, 12], this relation may be disturbed and therefore the computation is no longer valid. How can we explain the observed changes in renal hippurate extraction? In the present study EhP fell during placebo as well as
angiotensin II are comparable with previous findings on renal extraction of paraaminohippurate and orthoiodohippurate [2, 3], whereas the dopamine induced fall in EhI to our knowledge has enon which, in agreement with others [6], can be largely overcome never been reported. Several factors have been proposed to affect by correcting ERPFhP for Eh and Hct. However, the present renal extraction of hippurate or its derivatives [13]. First, a data also indicate that EhIP does not independently affect the pronounced fall in glomerular filtration rate (GFR) might induce relation between changes in ERPFhIP and changes in RBFprobe a fall in EhIP [14]. Second, EhIP might fall due to inhibition of the and can thus be substituted by a simple correction factor. This tubular transport mechanism by competition between drug and observation has an important clinical impact: the determination of hippurate for the organic anion transport mechanism [15—171. EhIP requires renal vein catheterization. Hct measurements, on Third, renal tracer extraction may be changed by alterations in the other hand, are relatively simple to perform. Since changes in intrarenal blood flow distribution [2, 5, 18, 19]. Fourth, due to this parameter independently affect the relation between changes changes in plasma transit time, EhP might change inversely with in ERPFhP and changes in RBFPrObe, changes in Hct must be changes in RBF [4, 5, 201. Lastly, it has been suggested that the established to obtain adequate results. Thus, our data show that rate of renal tracer secretion is enhanced by an increase in sodium
1622
Visscher et al: Hippurate clearance and renal blood flow
Selig showed, as we did, in conscious dogs using Doppler flow probes, that intravenous infusion of norepinephrine induces a rise
80
in RBF [32]. They attributed this RBF increase to a reflex mediated vasodilation, a mechanism which is likely to be inhibited during anaesthesia [33]. Another interesting example of anaesthesia/drug-effect interaction is the study of Swain et al [34] in dogs, who showed that during anaesthesia indomethacin induced a fall in RBF (probes), whereas during consciousness this fall was not
a)
C
a) U)
40
CO
0 U-
0
—40
—40
0
40
80
A RBFprobe (% of baseline)
Fig. 5. Correlation between drug-induced calculated changes in RBF (RBF1), based upon changes in ERPF,J, and changes in Hct, and changes in left renal blood flow measured by renal artely probe (RBF,.0b,.). Changes in response to enalaprilate (0), dopamine (•), norepinephrine (A), nitroprusside (h), indomethacin (Eli), and angiotensin II (•). The solid line indicates the line of identity, the dotted line indicates the regression line.
reabsorption rate [21, 22]. The observed fall in EhiP during placebo cannot be attributed to changes in renal function since
significant. In the present study, indomethacin indeed induced a non-significant fall in RBFPrObC. Thus, the present data once more stress that one should be careful when interpreting results from experiments during anaesthesia. In conclusion, the present data indicate that changes in plasma hippurate clearance are a good means to establish drug-induced relative changes in renal blood flow. A simple algorithm which takes changes in hematocrit into account suffices for the conversion of hippurate clearance data to changes in renal blood flow. If the algorithm is used, knowledge of the actual renal hippurate extraction is not needed. Since drug effects on plasma hippurate clearance, renal hippurate extraction, and hematocrit reported are comparable in dog and humans, extrapolation of the algorithm to humans appears valid. Therefore, renal vein catheterization to obtain renal hippurate extraction is unnecessary in this setting. Acknowledgments This study was made possible by grant (C92.126l) of the Dutch Kidney Foundation. We are indebted to Mr. J.M. Elstrodt for technical assistance, and to Mr. A. Nijmeijer for assistance in training the dogs.
Reprint requests to R.M. Huisman, M.D., Department of Medicine, RBF as well as RVR remained stable. Tubular hippurate secreof Nephrology, University Hospital, P.O. Box 30.001, 9700 RB tion, however, may have been inhibited by the glucose infusion Division Groningen, The Netherlands. either directly by competition [7] or indirectly by a fall in sodium reabsorption rate via extracellular volume expansion [22]. AnReferences other possibility is redistribution of renal blood flow in favor of the medulla due to renal nerve stimulation [23]. Both explanations are speculative. Changes in cortical transit time as a cause of 1. SMITH HW: The Kidney: Structure and Function in Health and Disease.
changes in EhIP appears a good overall explanation for the drug-induced changes in EhI, since in the present data a clear negative relation exists between changes in EhP and changes in RBFPrObC. Based upon the present results, however, it cannot be concluded what, if any, overall explanation is the correct one.
The hemodynamic effects of the studied vasoactive drugs compare well with previous data both in dog and man despite considerable differences in study design. As previously shown, ACEi induced a fall in systemic blood pressure (MAP) and renal vascular resistance (RVR), and a rise in RBF established either by
flow probe or by hippurate clearance [3, 24, 25]. The effects of angiotensin II were exactly the opposite [26]. Dopamine did not
affect MAP, but induced a fall in RVR and a rise in RBF independent of the technique used [27, 28]. In response to nitroprusside both MAP and RVR fell, whereas RBF, measured either way, did not change significantly [29, 301. The observation
that intravenously infused norepinephrine induced a rise in RBFProhe without a change in RVR, however, contrasts with the general opinion that norepinephrine induces a fall in RBF with an increase in RVR [31]. The latter conclusion is based on studies performed during anaesthesia and/or in which RBF was measured by means of clearance techniques. Indeed, Anderson, Korner and
New York, Oxford University Press, 1951 2. CARRIERE S, FRIBORO J: Intrarenal blood flow and PAH extraction during angiotensin infusion. Am J Physiol 217:1708—1715, 1969 3. WENTING GJ, TAN-TJONG HL, DERKX FHM, DE BRUYN JHB, MAN
IN 'T VELD AA, SCHALEKAMP MADH: Split renal function after captopril in unilateral renal artery stenosis. Br Med J 288:886—890, 1984 4. VELASQUEZ MT, NOTARGIACOMO AV, COHN JN: Influence of cortical
plasma transit-time on p-aminohippurate extraction during induced renal vasodilatation in anaesthetized dogs. C/in Sci 43:401—411, 1972 5. AUKLAND K, LØYNING EW: Intrarenal blood flow and para-aminohip-
purate (PAll) extraction. Acta Physiol Scand 79:95—108, 1970 6. LEVINSKY NG, LIEBERTHAL W: Clearance techniques, in Renal Physiology, Handbook of Physiology, section 8, edited by WINDHAGER EE,
New York, Oxford University PreSS, 1992, p 236 7. LOTE CJ, MCVICAR AJ, YARDLEY CP: Renal extraction and clearance
of p-aminohippurate during saline and dextrose infusion in the rat. J Physiol 363:303—313, 1985
8. ZWEENS J, SCHIPHOF P: Permanent catheterization of aorta and pulmonary artery in the dog. Pfldgers Arch 362:201—202, 1976 9. Ross J JR COVELL JW, FELD OK, SCHMID-SCI-IOENBEIN G: Cardio-
vascular system, in Best and Taylor's Physiological Basis of Medical Practice, edited by WEsT JB, Baltimore, Williams and Wilkins, 1990, pp 245—246
10. JONKER GJ, Dc ZEEUw D, HUISMAN RM, VAN DER HEM GK: Pharmacological zero for electromagnetic renal blood flow measurement. Pflugers Arch 403:220—221, 1985
Visscher et a!: Hippurate clearance and renal blood flow
1623
11. DONKER AJM, VAN DER HEM GK, SLUITER WJ, BEEKHUIS H: A
the intrarenal distribution of blood during the natriuresis accompa-
radioisotope method for simultaneous determination of the glomerular filtration rate and the effective renal plasma flow. Neth J Med
flying saline loading in the dog. J Clin Invest 44:929—941, 1965 23. POMERANZ BH, BIRTCH AG, BARGER AC: Neural control of intrarenat blood flow. Am J Physiol 215:1067—1081, 1968 24. HOLLENBERG NK, MEGGS LG, WILLIAMS GH, KATZ J, GARNIC JD, HARRINGTON DP: Sodium intake and renal responses to captopril in
20:97—103, 1977
12. VI5scHER CA, DE ZEEUW D, DE JUNG PE, PIERS DA, BEEKHUIS H,
HuISMAN RM. ACEI-induced changes in hippurate handling in renovascular hypertension. (abstract) Kidney mt 46:1472, 1994 13. MØLLER JV, SHEIKH MI: Renal organic anion transport system: pharmacological, physiological, and biochemical aspects. Pharmacol Rev 34:315—358, 1983 14. HABERLE DA, RUHLAND G, LAUSSER A, MOORE L, NEISS A: Influence
of glomerular filtration rate on renal PAH secretion rate in the rat kidney. Pflugers Arch 375:131—139, 1978 15. SINGHVI SM, DUCHIN KL, WILLARD DA, MCKINSTRY DN, MIGDALOF
BH: Renal handling of captopril: Effect of probenecid. Clin Pharmacol Ther 32:182—189, 1982
16. ULLRICH KJ, RUMRICH C, DAVID C, FRITZSCH G: Bisubstrates: Substances that interact with renal contraluminal organic anion and
cation transport systems. I. Amines, piperidines, piperazines, azepines, pyridines, quinolines, imidazoles, thiazoles, guanidines and hydrazines. Pflugers Arch 425:280—299, 1993 17. ULLRICH KJ, RUMRICH C, DAVID C, FRITZSCH G: Bisubstrates: Substances that interact with both, renal contraluminal organic anion and cation transport systems. II. Zwitterionic substrates: Dipeptides, cephalosporins, quinolone-carboxylate gyrase inhibitors and phospha-
mide thiazine carboxylates; nonionizable substrates: Steroid hor-
mones and cyclophosphamides. Pflugers Arch 425:300-312, 1993 18. PILKINGTON LA, BINDER R, DE HAAs JCM, Prrrs RF: Intrarenal distribution of blood flow. Am J Physiol 208:1107—1113, 1965
19. REUBI F: Objections a Ia théorie de Ia separation intrarbnale des hématies et du plasma (Pappenheimer). Helv Med Acta 4:516—523, 1958
20. CARRIERE 5, FRIBORG J, GUAY JP: Vasodilators, intrarenal blood flow, and natriuresis in the dog.AmJPhysiol 221:92—98, 1971 21. VOGEL G, KROGER W: Die Bedeutung des Transportes, der Konzen-
tration und der Darbietungsrichtung von Na für den tubularen Glucose- und PAH-transport. Plugers Arch 288:342—358, 1966 22. EARLY LE, FRIEDLER RM: Changes in renal blood flow and possibly
normal man and in essential hypertension. Kidney mt 20:240—245, 1981
25. ZIMMERMAN BG, GOERING JL, RAICH PC: Resistance to blockade by
saralasin of effect of ACE inhibitors in conscious sodium-restricted dogs. Am J Physiol 255:F944—F952, 1988 26. MITCHELL KD, NAVAR LG: Interactive effects of angiotensin II on renal haemodynamics and tubular reabsorptive function. Kidney mt 38(Suppl 30):69—73, 1990 27. MCDONALD RH JR, GOLDBERG LI, MCNAY JL: Effects of dopamine
in man; augmentation of sodium excretion, glomerular filtration rate and renal plasma flow. J Clin Invest 43:1116—1124, 1964 28. MCNAY JL, MCDONALD RH JR, GOLDBERG LI: Direct vasodilation produced by dopamine in the dog. Circ Res 16:510—517, 1965 29. NIELSEN CB, EISKJIER H, PEDERSEN EB: Enhanced renal production
of cyclic GMP and reduced free water clearance during sodium nitroprusside infusion in healthy man. Eur J Clin Invest 23:375—381, 1993 30. PAGANI M, VATNER SF, BRAUNWALD E: Hemodynamic effects of
intravenous sodium nitroprusside in the conscious dog. Circulation 57:144—151, 1978
31. SCHRIER RW: Effects of adrenergic nervous system and catecholamines on systemic and renal hemodynamics, sodium and water excretion and renin secretion. Kidney mt 6:291—306, 1974 32. ANDERSON WP, KORNER PT, SELIG SE: Mechanism involved in the renal responses to intravenous and renal artery infusions of noradrenaline in conscious dogs. J Physiol 321:21—30, 1981 33. VATNER SF, BRAUNWALD E: Cardiovascular control mechanisms in the conscious state. N Engi J Med 293:970—976, 1975 34. SWAIN JA, HEYNDRICKX GR, BOETFCHER DH, VATNER SF: Prosta-
glandin control of renal circulation in the unanesthetized dog and baboon. Am J Physiol 229:826—830, 1975