AMERICAN Vol. 230,

J~WRNAL No. 3, March

OF

PHYSIOLOGY 1.976. Printed

in U.S.A.

Effect of hemorrhagic shock on renal release of prostaglandin PAUL A. JOHNSTON AND EWALD E. SELKURT Department of Physiology, Indiana University Medical

JOHNSTON, PAUL A., AND EWALD E. SELKURT. Effect of hemorrhagic shock on renal release of prostaglandin E. Am. J. Physiol. 230(3): 831-838. 1976. -The effect of hemorrhage and reinfusion on renal release of prostaglandin E (PGE), arterial [PGE], mixed-venous [PGE], and renal function was observed in anesthetized dogs. Following hemorrhage to 60 mmHg arterial pressure, arterial [PGE] rose significantly from 405 to 740 pg/ml. Renal release of PGE remained near control (8 ng/min), as renal blood flow (RBF) decreased from 4.7 to 2.2 nl/min per gram kidney weight (K W). Mixed-venous [PGE] remained near the control value (960 pg/ml). Reinfusion of shed blood restored RBF to 4.0 ml/min per K W. Renal release of PGE rose significantly to 190 ng/min. Arterial [PGE] remained elevated, but mixed-venous [PGE] was not significantly different from control. Indomethacin, a prostaglandin synthesis inhibitor, caused a significant decrease in renal release of PGE. Arterial [PGE] remained elevated following treatment. The inhibition of PGE release from the kidney by indomethacin indicates that increased renal release of PGE following reinfusion is the result of accelerated PGE synthesis. The data suggest that the elevated arterial [PGE] may be the result of alteration of the handling of PGE by the lung.

renal PGE synthesis; renal hemodynamics; handling in shock; indomethacin

renal

ROLE

Center,

Indianapolis,

Indiana

46202

posed. An important consequence of an increase in PGE synthesis, as mentioned previously, might be an increase in the systemic arterial [PGE], thus enhancing the vasodilator influence on systemic blood pressure. Third, various agents, release of which is enhanced in hemorrhage and which affect renal function, i.e., angiotensin, catecholamines, bradykinin, sympathetic nerve activity, and changes in blood flow and pressure have been shown to increase the renal venous concentration of PGE (9, 10, 13, 14, 16, 28-30, 36, 46, 47). Some of these agents, such as angiotensin and the catecholamines, are elevated during hypotension and postreinfusion normovolemic shock (8, 22, 36, 41, 42). Finally, studies involving the infusion of prostaglandins (PGE, PGA) into the renal artery (4, 7, 19, 20, 43) have described effects on renal function that are very similar to changes occurring in normovolemic shock (38-40, 48). The present investigation was designed to quantitate renal release of PGE during periods of prolonged hypotension due to hemorrhage and following reinfusion of shed blood. An attempt was made, aided by the action of the drug indomethacin, to define the effect of a change in renal PGE production on the systemic arterial [PGE], mixed-venous [PGE], and in turn, renal hemodynamics and electrolyte and water handling.

electrolyte

in the control of renal function, and as a renal factor contributing to the control of arterial blood pressure in the anesthetized dog has been the object of much attention (4, 7, 19, 20, 24, 28). Recent evidence suggests that accelerated prostaglandin production by the kidney might occur during the hypotensive phase or the normovolemic phase of hemorrhagic shock (9, 26). This event might, in turn, affect the level of arterial blood pressure, and renal function, since the prostaglandins have been shown to alter significantly renal hemodynamics, electrolyte, and water handling, Several lines of evidence give credence to this suggestion. First, the kidney of several species has been shown to possess the capacity to synthesize and break down the prostaglandins (6, 11, 18, 31). Second, studies involving hypotension produced by hemorrhage and the administration of endotoxin from Escherichia coli have demonstrated changes in the renal venous prostaglandin E (PGE) concentration during the hypotensive periods (9, 26). The implication of these studies is that there was an increased synthesis of PGE under the conditions im-

THE

E

METHODS

OF the prostaglandins

This study was performed on 26 male mongrel dogs weighing 13-28.5 kg. The animals were anesthetized with sodium pentobarbital, 30 mg/kg in 0.9% saline. Anesthesia was maintained with periodic administration of sodium pentobarbital. The preoperative procedure included removal of food 18-24 h prior to the initiation of surgery. The animals were allowed free access to milk and water during this period. Both the right and left femoral artery and vein were isolated and catheterized. Mean arterial blood pressure (MABP) was measured through the right femoral arterial catheter with a Statham P23Db pressure transducer and a Beckman recorder. Saline solution (0.9%) containand para-aminohippurate ing inulin or creatinine (PAH) was infused through the right venous catheter at a minimal rate (0.2 ml/min) so as not to disturb the fluid balance. A 25ml priming solution of PAH, inulin, or creatinine in 0.9% saline was injected intravenously to establish plasma levels of PAH averaging ca. 2 mg/lOO ml; inulin, 25 mg/lOO ml; creatinine, 20 mg/lOO ml. Catheters were placed in the left femoral artery and 831

Downloaded from www.physiology.org/journal/ajplegacy at Tulane University (129.081.226.078) on February 15, 2019.

832

P. A. JOHNSTON

vein for withdrawing and reinfusing blood. The blood shed to reduce arterial pressure was held in a polyethylene reservoir bottle at a temperature of 34-36°C for the duration of the hypotensive period (X0-180 min). The left renal vessels were isolated, retroperitoneally, through a flank incision. A catheter was placed in the ureter, and a fishhook needle (18 gauge) was placed in the left renal vein to sample the effluent. Alternatively, the spermatic vein was catheterized, with the catheter tip placed at the junction of the renal vein. Left renal blood flow was measured with an electromagnetic flowmeter probe (Carolina Medical Electronics) placed on the left renal artery. In several animals, a cardiac catheter was placed in the right ventricle in order to sample mixed-venous blood prior to its transit through the lung bed. One group (n = 8) served as a nonhemorrhaged control in which the same functional parameters were followed as in the experimental group over a comparable time period (10 h). The remaining 18 animals were arranged into two experimental subgroups: one untreated (group I); the other, indomethacin treated (group 11). Following two control collection periods (30 min each), mean arterial blood pressure was reduced in both groups to 60 mmHg by hemorrhage of 2.5% body wt (range of initial bled volumes was 250-760 ml) and maintained for 90 min by small i .ncrements into the reservoir bottle. Following the 60 mmHg hypotensive phase, blood pressure was further lowered to 30-40 mmHg by total hemorrhage of 3.2% body wt (the range of total bled volumes was 425-850). The blood pressure was stabilized by continuity of the arterial catheter with the blood in the reservoir held at a level to maintain the desired pressures. One 60-min urine collection period was then taken, but in the majority of cases there was no urine flow during this time and, consequently, only blood samples were taken. Blood was reinfused through the left femoral vein over a period of 30 min. A further 30- to 45-min period of time elapsed before collection of urine was begun. This insured adequate flushing of stagnant urine from the kidney. Urine was collected throughout the postreinfusion phase in periods of 30 min. Total postreinfusion observation time was 5 h. Group II animals (n = 8) were treated with indomethacin, dissolved in 0.1 M phosphate buffer solution to give a final concentration of 10 mg/ml. The drug was given in two injections of 2 mg/kg iv after 90 min and 210 min of postreinfusion monitoring. Group I animals were monitored through hypotension and postreinfusion without receiving drug treatment (n = 10). Plasma and urine samples were taken for the measurement of concentrations of sodium, calcium, phosphate, inulin, PAH, total osmolality, and PGE. Total urinary and plasma sodium and calcium were measured by atomic-absorption spectrophotometry. Plasma ionized calcium was determined with an ion-selective electrode. An average value of 50% plasma ionized calcium, which remained relatively constant throughout control, hypotension, and postreinfusion, was determined from ionized calcium measurements and total plasma calcium. Inorganic phosphate was determined by the

AND

E. E. SELKURT

method of Gomori (17). The anthrone method of Davidson and Sackner (12) was used to measure plasma and urine concentrations of inulin. Creatinine was measured in urine and trichloroacetic acid-plasma filtrate by the method of Kennedy et al. (25). The concentration of PAH was measured by the method of Smith. et al. (44), and total osmolality was measured by freezing-point depression on an Advanced Instruments osmometer. Arterial and venous whole-blood samples, drawn for the analysis of PGE, were immediat.ely centrifuged at 4°C for 20 min. Plasma from these samples was frozen at -20°C and extraction of 1 ml of plasma was carried out within 24 h of the initial sampling. The method for extraction and assay of prostaglandins was essentially that employed by Jaffe and Behrman (23). The prostaglandins were extracted from raw plasma by washing with an equal volume of acidic ethyl acetate. Tritiated PGE was added to l-ml samples of dog plasma and extracted along with the experimental samples. The PGEs were separated on silica-gel columns. The efficiency of the extraction procedure was determined by comparing the counts per minute in the fraction collected from the l-ml aliquots of dog plasma to the counts per minute of the prostaglandin added initially. Overall recovery of PGE was 85.8%. Column separation showed that only 2.4% of PGA appeared in the fraction. Data for PGE only will be given consideration in this presentation. The amount of PGE in each sample was quantitated using the double-antibody radioimmunoassay developed commercially by Clinical Assays Incorporated. Tne renal release of PGE was calculated by the following formula release

(ng/min)

=

] x venous blood flow) ([PGEvenous - ([PGEarteriall x arterial

blood flow)

Venous outflow differs from arterial inflow by the amount of urine flow. Maximum urine flow in any single animal was approximately 1 ml/min. This was a fraction of a percent (cl%) of the total blood flow to the left kidney. For this reason, arterial blood flow to the left kidney was not corrected for urine flow to determine venous outflow, and arterial blood flow and venous blood flow were assumed to be equal. The above equation then simplifies to release

([PGL,,d

(ng/min)

=

- [PGEarteriall) x total renal

plasma

flow to left kidney

where [PGE] represents the concentration in plasma. The fractional clearance of electrolytes and water was determined by comparing their clearances as a ratio to that of inulin or creatinine, which measured glomerular filtration rate (GFR). Renal vascular resistance (RVR) of the left kidney was calculated as MABP/RBF. The Student t tests for equal and unequal sample size were applied to the data to determine the significance of the changes within a group of animals, and between the

Downloaded from www.physiology.org/journal/ajplegacy at Tulane University (129.081.226.078) on February 15, 2019.

RENAL

RELEASE

OF PROSTAGLANDIN

E AND

HEMORRHAGIC

833

SHOCK

group of animals which was hemorrhaged and reinfused and that group which was hemorrhaged, reinfused, and treated with indomethacin following the reinfusion.

6

RESULTS 0

ControZ observations. Control observations are summarized in Figs. 1 and 2. None of the hemodynamic parameters showed a significant change over the 10-h period (Fig. 1). Nor did arterial plasma PGE concentration and renal PGE release vary significantly during the period of observation. With regard to data on electrolyte and water handling, fCca, fir, and fCN, showed downward trends. The latter two functions were statistically significant (P = .05) or approached significance in periods 8, 9, and 10. fC,,M and (P = .05-.lO) fT & remained reasonably constant throughout. and reinfusion on renal hemoEffect of hemorrhage dynamics, PGE release, arterial [PGE], and renal function. Mean control renal release of PGE was 8 ng/min; lower than the control group, but not statistically so. Each animal showed good stability of this value through l-2 h of control. Mean arterial [PGE] remained relatively stable during control collections at 405 pg/ml, again fortuitously lower than the control group (P < .05) Figure 3 demonstrates the effect of hemorrhage and reinfusion on renal hemodynamics, arterial [PGE], and the release of PGE from the left kidney. A significant

1

i

*

t

T

2 500

--L

T

T

ARTERIAL

RENAL

1

6 m z k-f 4 : c2 t!

0

I

I

I

I

I

I

1

I

I

I

1

2

3

4

5

6

7

8

9

10

1

(HOURS)

FIG. 2. Control data on fractional urine volume (fir), and fractional osmolar, sodium, and calcium clearance. f&o, (not shown) was done only in 4 animals and showed no change. (Means + SE.)

I

MABP

RVR

t

i

PGE

PGE

C:NC

RELEASE

40

5

2

5

7

9

(HOC’RS)

FIG. 1. Control hemodynamic data in 8 dogs observed over a 10-h period, corresponding to duration of observation in hemorrhagic shock series. Effects on arterial [PGE] and renal PGE release are also shown. For method of calculation of RVR (renal vascular resistance), and renal PGE release, see text. (Means + SE.)

rise in renal vascular resistance (RVR) (P < 0.05) and a significant fall in RBF (P < 0.001) accompanied the decrease in blood pressure to 60 mmHg. A further fall (P < 0.005) in RBF and rise (P < 0.001) in RVR were observed with a decrease in blood pressure to 30-40 mmHg. During the 60 mmHg hypotensive period there was a statistically significant rise in arterial [PGE] (P < 0.025). Renal release of PGE rose slightly, early in the period, but this was not statistically significant. However, at this time renal venous [PGE] actually increased from an average of 461.9 to 625.6 pg/ml. Thus the insignificant increase in release reflects the major influence of the reduction in RBF and uneven perfusion of the medullary sites of synthesis of PGE in hemorrhage. Following reinfusion, RVR and RBF returned to values that were not significantly different from the initial control. MABP returned to 100 mmHg, compared to the control value of 113 mmHg, and slowly declined to 90 mm by the end of the postreinfusion monitoring period. Reinfusion was accompanied by a significant increase in renal PGE release (P < 0.010). This release remained elevated throughout the postreinfusion observation period. Arterial [PGE] decreased slightly below the hypotensive period level, but showed an increasing trend toward the end of the reinfusion period. Data relating mixed-venous [PGE] to arterial [PGE] and renal PGE release are also presented in Table 1. There was a slight increase in mixed-venous [PGE] following hemorrhage to 60 mmHg. This elevation did not persist during hypotension and was not statistically

Downloaded from www.physiology.org/journal/ajplegacy at Tulane University (129.081.226.078) on February 15, 2019.

834

P. A. JOHNSTON t

1

’

I

I

1

.

.

I

1

I

11

I

I

.

I

11

120

80

0

4 I M

60

3 ‘=;E

40

\

g E E

20 1 7.0

.z $‘r; E

I

n=ll I

I

n=5

.

1

.

i

R’BF

1 I

5.0 3.0 1.0 I

1000

-

.

a

1

1

ARTERLAL

PGE

.

.

.

l

.

.

1

1

I

I

I

4

5

6

1s

1

L

.

I

I

I

I

I

7

8

9

10

1.

CONC.

2 800 aM 600

RENAL

PGE

RELEASE

I

I 0

I

I

1

1

2

3

J

(HOURS) FIG. 3. Effect of hemorrhage and reinfusion on renal hemodynamics, renal release of PGE, and arterial [PGE]. Data through first 7 h of observation were collected from 11 animals (but only 10 animals in case of PGE output and arterial [PGE]). Remainder of data (to right of vertical dashed line) are collected from 5 animals which survived longer. Means plotted are overall control means, plus mean difference from control of each collection period. Standard errors are standard error of mean difference from control at each collection period.

significant when compared to the control. Following reinfusion, there was an initial rise in mixed-venous [PGE] to 42.75% greater than control; however, this was not statistically significant. Throughout the latter half of the postreinfusion observation period, mixed-venous [PGE] declined. Correspondingly, arterial [PGE] rose significantly during hypotension, and remained elevated with respect to control following reinfusion, particularly toward the end of the observation period. It should be noted that the increase in arterial [PGE] toward the end of the observation period occurred despite a decrease in mixed-venous [PGE]. This seemed to indicate an increased passage of PGE across the lung bed. In several instances, the shed blood was sampled with

AND

E. E. SELKURT

time to determine if a net production of PGE had taken place during the period of exsanguination. Samples taken 30 min after hemorrhage averaged 784 t 49 pg/ml aver(n = 4). Those taken at 90 min after hemorrhage aged 674 t 61 pg/ml (n = 4). It is evident from these data that no net PGE production took place in the reservoir of exsanguinated blood. Data describing the effect of hemorrhage and reinfusion on fractional sodium (fc,,), calcium (fC&, and phosphate (fC,,,) clearance from the left kidney are presented in Fig. 4. No data are presented for the 30-40 mmHg phase of hypotension because of the difficulty in acquiring urine samples of adequate volume. A significant decrease in both the fCN, (P < 0.05), and fCpo4 (P < 0.025) can be seen during the later half of the 60 mmHg phase of hypotension. The fCca was increased throughout the 60 mmHg phase of hypotension (P < 0.001). Following reinfusion, fCN, and fCca were elevated through the first 2 h of observation (P < 0.05). These values then returned to control. The fCpo4 remained elevated throughout the entire postreinfusion period (P < 0.005), but was decreasing during the 9th and 10th h. These increases contrast markedly with the downward trend for fCNa and fCca observed throughout in the control series (Fig. 2). Figure 5 is a presentation of the responses of fractional negative free water clearance (f T &o>, fractional osmolar clearance (fCoSM), fractional urine flow rate (fv), and GFR/lOO g KW to hemorrhage and reinfusion. Hemorrhaging to 60 mmHg caused no significant change in fCoSM of fT&o. However, a significant rise occurred in fir (P < O.OOl), and significant fall occurred in GFR/lOO g KW (P < 0.001). Following reinfusion, were elevated. The elevation f% fGsM, and fT&o persisted through the balance of the postreinfusion period. Again, control group fV fell, and fT$& showed no significant change at this time. GFR/lOO g KW remained significantly depressed throughout the postreinfusion phase, contrasting to the control series, which showed a slight increase (Fig. 1). Effect of indomethacin treatment following reinfusion. Indomethacin, an inhibitor of prostaglandin synthesis, was administered to the animals in group II after three urine collections had been made in the postreinfusion period (2 mg/kg iv). Figures 6 and 7 depict the response of the renal release of PGE, arterial [PGE], and hemodynamics to the indomethacin treatment. FigTABLE 1. Comparison of response of renal release of PGE, mixed-venous [PGE], and arterial [PGEI to hemorrhage and reinfusion Time,

O-60 90-180 210-270 305-365 365-445 445-505 525-585

min

MABP, mmHg

Renal Release of PGE, ng/min per 100 g KW

113 60 30 107

9213 15255 4225 120233

103 97 93

173+30 140+30 187241

Data are means + SE for the period of time 6. All other data are derived from Fig. 3.

Mixed-Venous

[PGEI, m/ml

9602 150 1,010~173 9632210

405+53 720+:70

1,070?200 1,250+215 9922183 800+161 indicated.

Arterial

[PGEI, pg/ml

For

mixed-venous

Downloaded from www.physiology.org/journal/ajplegacy at Tulane University (129.081.226.078) on February 15, 2019.

660283 600+40 500243 520249 660263 blood

PGE,

n =

RENAL

RELEASE

OF PROSTAGLANDIN

E AND

HEMORRHAGIC

The influence of indomethacin on the variables describing electrolyte and water handling by the kidney are presented in Fig. 8. There was a significant increase in fC& following the second injection of indomethacin (P < 0.005). Of interest is the observation the fCCa declined significantly after postinfusion period 4 (P < .OZ), while fCpo4 remained elevated. A significant decline in fCN, (P < 0.025) was noted following both injections of the drug. No significant changes could be demonstrated for the other variables, when compared to changes taking place in a corresponding time period in the nontreated animals.

120

if Ii E

80 40

6.5

m 2 +x si u-

5.5

4.5

3,5

2.5

I

I

80

835

SHOCK

I

n=18

I

n=lO

I I TT

n=17

I

n=9

DISCUSSION

This study presents evidence that PGE release from the kidney was significantly increased following reinfusion and remained elevated throughout the postreinfusion period. The fall in this release upon treatment with indomethacin would indicate that the increased release

3

80

E

I I

40 n=18

1

n=lO

I

I

5 HOURS

of hemorrhage and reinfusion on fCCa, fCpo4, and of each variable through the 3rd postreinfusion fGa. Responses period (left of vertical dashed line) are lumped means and SEs of animals in both group I and group II (n = 17 and 18). Thereafter, responses shown are only in group I animals (n = 9 and 10).

n=18

1

n=lO

n=18

! I--

n=lO

FIG. 4. Effect

80 3 t&l

2

ure 6 shows a significant fall in the renal release of PGE following indomethacin administration (P < 0.005). Accompanying the fall in PGE release was a fall in RBF in the treated group from 4.56 ml/min per g K W to 3.83 ml/ min per g K W (P < 0.001) (Fig. 7). RBF fell still further after a second injection of the drug, at 210 min after reinfusion. This change, however, was not significant. RVR increased significantly (P < 0.010) following the initial injection of the drug, from 26.44 to 35.63 mmHg/ ml per min per g K W. No further rise in RVR was observed following the second injection of indomethacin (Fig. 7). GFR decreased only after the second injection (periods 7 and 8). Very little change was noticed in arterial [PGE] in group II following administration of the drug. The rise in arterial - [PGE] prior to and following the second injection of indomethacin parallels a similar rise in the nontreated group (Fig. 3). No statistical significance could be demonstrated between the two groups. Therefore, no effect on arterial [PGEj could be attributed to the drug. Arterial [PGE] apparently rose despite treatment with indomethacin,

60

5 40

k.s 20

-I as XB g 2 Rlx

l T -

5 3 1 n=18

3 -I

O

UZ *

I

2

I 1

1 0

1

2

3

4

5

6

7

8

9

10

HOURS

of hemorrhage and reinfusion on urine flow, GFR, [CosM,and fT” HzO- To left of vertical dashed line (72 = IS>, lumped means of animals in groups I and II. To right (n = lo), responses of group I animals only, FIG.

5. Effect

Downloaded from www.physiology.org/journal/ajplegacy at Tulane University (129.081.226.078) on February 15, 2019.

836

P. A. JOHNSTON

El

1200

IASE D& WQ e

800

%

400

200

prehemorrhage

--IT

T

m

indomethacin-treated

m

indomethacin-treated preinjection

&

100 prehemorrhage

POSTREINFUSION

COLLECTION

PERIOD

6. Effect of indomethacin injection (2 mg/kg iv) on renal PGE release and arterial [PGE] in the postreinfusion phase. Dark bars represent mean (* SE) response of indomethacin-treated group of animals. Cross-hatched bar represents mean control response for same group prior to indomethacin treatment. Dotted horizontal line represents prehemorrhage control value. FIG.

_ 150

MABP

z cu

3

-----------

; 100

prehemorrhage control

50

5 63

3

‘2; E

l

1 RBF

---------__ prehemorrhage control

AND

E. E. SELKURT

that an increased renal release of PGE, as measured, indicates an increased synthesis of PGE by the kidney under the conditions of this study. The exact mechanism responsible for the increased production of renal PGE cannot be defined here. Angiotensin and the catecholamines have been shown to increase the concentration of PGE in renal venous blood (13, 16, 22, 28). Since these substances have likewise been demonstrated to be elevated in hypotension and postreinfusion normovolemic shock (22, 26, 36, 42, 46), they can be viewed as candidates for stimulation of It is also possible that inprostaglandin synthesis. creased levels of catecholamines and angiotensin in the shed blood, when reinfused into the system, may have provided a stimulus for renal PGE production. No attempt was made to measure these in the shed blood. As suggested previously, important supportive evidence for these data has recently come from the observations of Collier et al. (9) and Kessler et al. (26). These workers have shown the appearance of an increased amount of PGE-like material in renal venous blood following administration of the endotoxin from E. cob and hypotension due to hemorrhage. An important consideration with these data is that if the experimental procedures reduced blood flow, and this is certainly the expectation in either method of producing shock, renal venous concentrations of the prostaglandins may be increased without a change in synthesis. In the present study, renal venous concentrations of PGE were also elevated during hypotension, but when output was calculated in the presence of reduced RBF as described previously, no signficant change could be demonstrated. That the increased production of PGE during the normovolemic stage of shock may be closely related to renal hemodynamics has been pointed out. The admin-

\ $y&

40

&i EE E

20

-------*-

5.

^^

s& b” 8 -y I ‘=E

1 GFR

-

prehemorrhage control

e-----e----

t

T

prehemorrhage

.

40

‘1

20

-L

6

ml 0

POSTREINFUSION

COLLECTION

4

PERIOD

xB

7. Effect of indomethacin injection (2 mg/kg iv) on renal hemodynamics and MABP in the postreinfusion phase. Dark bars represent mean (* SE) response of indomethacin-treated group. Cross-hatched bar is mean response of indomethacin-treated group prior to indomethacin treatment (first 3 periods of posttransfusion phase).

4

FIG.

was due to a greater rate of synthesis, not simply increased washout from the kidney. One must realize that changes in renal venous release may reflect only part of the synthesis of PGE by the kidney. It is probable that a substantial amount of prostaglandin may be excreted as PGE in the urine, as well as its metabolites in both urine and venous blood. Neither the PGE in urine nor the metabolites in urine or venous blood were measured in this studv. Nevertheless. it is reasonable to assume

2

2

-------

POSTREINFUSION

COLLECTION

PERIOD

FIG. 8. Effect of indomethacin on fCNa, fCpol, and fCca in postreinfusion period. Dark bars represent mean (+ SE) response of indomethacin-treated group. Cross-hatched bar represents mean response of same group prior to indomethacin treatment.

Downloaded from www.physiology.org/journal/ajplegacy at Tulane University (129.081.226.078) on February 15, 2019.

RENAL

RELEASE

OF PROSTAGLANDIN

E AND

HEMORRHAGIC

istration of indomethacin caused a substantial increase in RVR and a decline in RBF (Fig. 7). Accompanying these changes was a significant decline in PGE release by the kidney (Fig. 6). RBF fell by 16.7% after administration of the drug, and RVR increased by 28.1%. One might speculate that, with the increased production of PGE by the kidney in the postreinfusion phase, RBF would be more dependent on PGE, and would be more greatly affected by the cessation of PGE production. Ultimately the control of renal blood flow in the postreinfusion period is dependent not only on the vasodilator influences, but on the vasoconstrictor influences as well. Thus the constrictor/dilator ratio may be slightly increased in postreinfusion shock, and slightly depressed renal blood flow is evidence for this. Conceivably, the increased PGE production by the kidney was helpful to the organ to the extent that it maintained blood flow to the kidney in the face of an overall increase in systemic humoral vasoconstrictor influences. An important consequence of hemorrhage and reinfusion was the development of increased arterial [PGE] during hypotension, and the persistence of this increase following reinfusion. A possible source of this increase is the increased renal release of PGE upon reinfusion. Two ideas are germane to this point. First, arterial [PGE] rose during hypotension when there was no increase in renal release. Second, the increased renal release after reinfusion failed to alter significantly the mixed-venous [PGE]. If one calculates the contribution of renal PGE production to mixed-venous [PGE], using 2.0 liters/min for cardiac output (41), 120 g total KW for an average sized animal used in these studies, and 140 ng/min per 100 g KW for renal PGE production in the postreinfusion period, one finds that the kidneys supply less than 10% of the mixed-venous [PGE]. This value is clearly within the variability we observed in mixed-venous [PGE] with the radioimmunoassay technique. Increased shunting through the lungs remains as a likely possible mechanism for the increase in arterial [PGE]. Several studies have suggested that parts of the lung bed become clogged with cellular debris during hemorrhage allowing for considerable shunting of blood flow (3, 5, 21, 34) through zones where metabolism of PGE is low, or because of bypassing such sites. Therefore, it seems very possible that the elevated arterial [PGE] during hypotension and following reinfusion is the result of an increased passage of PGE across the pulmonary vascular bed. Alternatively, the lung may itself contribute prostaglandin in increased amounts as a result of hemorrhagic shock. One must, of course, consider the production of PGE by the organ systems distal to the lung and proximal to the arterial blood collection site as possible sources for the increased arterial [PGE] (45). Blood platelets may be possible candidates for production of PGE under the conditions of the present study. It should be recalled, however, that samples of reservoir blood

837

SHOCK

taken periodically during the period of exsanguination showed no production of PGE. The effect of hemorrhage and reinfusion on electrolytes and water handling is well documented for both dog and primate (37-39,40) and serve, in this study, as a reference to which the effect of indomethacin on postreinfusion renal function can be compared. Small changes in fCCa, fCNa, and fCpo4 were noted after treatment with indomethacin. These may indicate a role for PGE in postreinfusion renal function. However, the changes may be the result of a more direct effect of indomethacin on renal function. Indomethacin has been shown to be an inhibitor of phosphodiesterase activity (15), and by this mechanism can effectively increase tissue cyclic AMP. Since cyclic AMP plays some role in renal function, especially with regard to calcium, phosphate, and possibly sodium handling (1, 2, 32, 33), indomethacin may be exerting an effect by this means as well as through inhibition of PGE synthesis. A postulated increased influence of cyclic AMP would tend to increase sodium and calcium reabsorption and reduce phosphate reabsorption (35). In summary, we have presented evidence that hypotension resulting from hemorrhage caused a significant rise in arterial [PGE], which remained significantly Since mixed-venous elevated following reinfusion. [PGE] did not rise significantly either during hypotension or following reinfusion despite increased renal release, it seems reasonable to suggest that the increased arterial [PGE] was the result of an alteration in pulmonary handling of PGE following the hemorrhagic stress. Renal release of PGE was not significantly elevated during hypotension, but rose following reinfusion and remained elevated throughout the postreinfusion period. The mechanism of increased PG synthesis is not known, but elevation of plasma angiotensin and catecholamines could be the basis. This investigation supplies no direct evidence that increased arterial [PGE] or increased renal synthesis of PGE had any role in renal functional changes (electrolyte and water handling) during postreinfusion. However, assuming indomethatin has no direct vasoconstrictor effect on the vasculature, modulation of renal blood flow and vascular resistance in the shocked animal seemed to be dependent on PGE release (27). An intravenous role needs yet to be firmly established, also. We thank Dr. Ronald R. Beck for his knowledgeable criticism and discussion of this manuscript. We are also grateful for the technical assistance of Christine Amneus, Cynthia Blasingham, Carol Robideau, Carol Van Ermen, and Mary Ann Neel. This study was supported by National Science Foundation Grant GB 43372. Portions of this work were presented to the Federation of American Societies for Experimental Biology (Johnston, P. A., and R. R. Beck. Effect of hemorrhage and reinfusion on renal output and systemic arterial concentration of prostaglandin E (Abstract). Federation Proc. 33: 348, 1974). Received

for publication

18 July

1975.

REFERENCES 1. AGUS, Z., L. B. GARDNER, of parathormone on renal dium, and phosphate. Am. 2. AGUS, Z., J. B. PUSCHETT,

L. H. BECK, AND M. GOLDBERG. Effect tubular reabsorption of calcium, soJ. Physiol. 224: 1143-1148, 1974. D. SENESKY, AND M. GOLDBERG. Mode

of action of PTH and cyclic AMP on renal tubular phosphate reabsorption in dog. J. CZin. Invest. 50: 617-626, 1971. 3. ALLARDYCE, B., H. F. HAMIT, T. MATSUMOTO, AND R. V. MOSELEY. Pulmonary vascular changes in hypovolemic shock: radiog-

Downloaded from www.physiology.org/journal/ajplegacy at Tulane University (129.081.226.078) on February 15, 2019.

P. A. ,JOHNSTON

838

4. 5.

6. 7.

8.

9.

10.

11. 12. 13.

14.

15. 16.

17.

18. 19. 20.

21.

22.

23.

24. 25.

26.

27.

raphy of pulmonary microcirculation and the possible role of platelet embolism in increasing vascular resistance. J. Trauma 9: 403-411, 1969. ARENDSHORST, W. J., P. A. JOHNSTON, AND E. E. SELKURT. Effect of prostaglandin E, on renal hemodynamics in nondiuretic and volume-expanded dogs. Am. J. Physiol. 226: 218-225, 1974. AYERS, S. M., H. MUELLER, S. GIANELLI, P. FLEMING, AND W. J. GRACE. The lung in shock. Alveolar-capillary gas exchange in the shock syndrome. Am. J. CardioZ. 26:‘ 588-594, 1970. BERGSTROM, S. Prostaglandins: members of a new hormonal system. Science 157: 382-391, 1967. BERL, T., AND R. W. SCHRIER. Mechanism of effect of prostaglandin E, on renal water excretion. J. CZin. Invest. 52: 463-471, 1973. CLAYBAUGH, J. R., AND L. SHARE. Vasopressin, renin, and cardiovascular responses to continuous slow hemorrhage. Am. J. PhysioZ. 224: 519-523, 1973. COLLIER, J. G., A. G. HERMAN, AND J. R. VANE. Appearance of prostaglandins in renal venous blood of dogs in response to acute systemic hypotension produced by bleeding or endotoxin. J. Physiol., London 230: 19P-2OP, 1973. CROWSHAW, K., J. C. MCGIFF, N. A. TERRAGNO, A. J. LONIGRO, M. A. WILLIAMSON, J. C. STRAND, J. B. LEE, AND K. K. F. NG. Prostaglandin-like substances present in blood during renal ischemia: patterns of release and their partial characterization. J. Lab. CZin. Med. 74: 866, 1969. CROWSHAW, K., AND J. Z. SZLYK. Distribution of prostaglandins in rabbit kidney. Biochem. J. 116: 421-424, 1970. DAVIDSON, W. D., AND M. A. SACKNER. Simplification of the anthrone method for the determination of inulin in clearance studies. J. Lab. CZin. Med. 62: 351-356, 1963. DUNHAM, E. W., AND B. G. ZIMMERMAN. Release of prostaglandin-like material from dog kidney during nerve stimulation. Am. J. PhysioZ. 219: 1279-1285, 1970. EDWARDS, W. G. JR., C. G. STRONG, AND J. C. HUNT. A vasodepressor lipid resembling prostaglandin E, in renal venous blood of hypertensive patients. J. Lab. CZin. Med. 74: 389-399, 1969. FLORES, A., AND G. W. G. SHARB. Endogenous prostaglandins and osmotic water flow in the toad bladder. Am. J. Physiol. 223: 13924397, 1972. FUGIMOTO, S., AND M. F. LOCKETT. The diuretic action of prostaglandin E, and of noradrenalin and the occurrence of a prostaglandin E-like substance in the renal lymph of cats. J. Physiol., London 208: 1-19, 1970. GOMORI, G. A modification of the calorimetric phosphate determination for use with the photoelectric calorimeter. J. Lab. CZin. Med. 27: 955-960, 1942. HAMBERG, M. Biosynthesis of prostaglandins in the renal medulla of rabbit. FEBS (Fed. European Biochem. Sot.) Lett. 5: 127-130, 1969. HERZOG, J., H. JOHNSTON, AND D. P. LAULER. Natriuretic effect of PGE, in dog kidney. CZin. Res. 14: 491, 1966. HERZOG, J. H., H. JOHNSTON, AND D. P. LAULER. Effects of sodium, prostaglandins E,, EZ, and A, on renal hemodynamics, and water excretion in the dog. CZin. Res. 16: 386, 1968. HILLEN, G. P., W. D. GAISFORD, AND C. G. JENSEN. Pulmonary changes in treated and untreated hemorrhagic shock. Am. J. Surg. 122: 639-649, 1971. HODGE, R. L., R. D. LOWE, AND J. R. VANE. The effect of alteration of blood volume on the concentration of circulating angiotensin in anesthetized dog. J. Physiol., London 185: 613-626, 1966. JAFFEE, B. M., AND H. R. BEHRMAN. Prostaglandins E, A, and F. In: Methods of Hormone Radioassay, edited by B. M. Jaffe and H. R. Behrman. New York: Academic, 1974, p. 19-34. JOHNSTON, H. H., J. H. HERZOG, AND D. P. LAULER. Effect of prostaglandin E, on renal hemodynamics, sodium, and water excretion. Am. J. PhysioZ. 213: 939-946, 1967. KENNEDY, T. J., JR., S. G. HILTON, AND R. W. BERLINER. Comparison of inulin and creatinine clearance in the normal dog. Am. J. PhysioZ. 171: 164-168, 1952. KESSLER, E., R. C. HUGHES, E. N. BENNETT, AND S. Y. NADELA. Evidence for the presence of prostaglandin-like material in the plasma of dogs with endotoxin shock. J. Lab. CZin. Med. 81: 8594, 1973. LONIGRO, A. J., H. ITSKOVITZ, K. CROWSHAW, AND J. C. MCGIFF.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37. 38. 39.

40.

41. 42.

43.

44.

45.

46.

47. 48.

AND

E. E. SELKURT

Dependency of renal blood flow on prostaglandin synthesis in the dog. CircuZation Res. 32: 712-717, 1973. MCGLIFF, J. C., K. CROWSHAW, N. A. TERRAGNO, AND A. J. LONIGRO. Release of a prostaglandin-like substance into renal venous blood in response to angiotensin II. CircuZation Res. 27, Suppl. I: 1-121-I-130, 1970. MCGIFF, J. C., K. CROWSHAW, N. A. TERRAGNO, A. A. LONIGRO, J. C. STRAND, F. A. WILLIAMSON, J. B. LEE, AND K. K. F. NG. Prostaglandin-like substances appearing in canine renal venous blood during renal ischemia. Their partial characterization by pharmacological and chromatographic procedures. CircuZation Res. 27: 765-782, 1970. MCGIFF, J. C., N. A. TERRAGNO, K. U. MALIK, AND A. J. LONIGRO. Release of a prostaglandin-like substance from canine kidney by bradykinin: comparison with Eledoisin. CircuZation Res. 31: 36-43, 1972. MUIRHEAD, E. E., B. BROOKES, AND J. A. PITCOCK. Renomedullary antihypertensive function in accelerated hypertension. Observations on renomedullary interstitial cells. J. CZin. Invest. 51: 181-190, 1972. ORCZYK, G. D., AND H. R. BEHRMAN. Ovulation blockade by aspirin or indomethacin in vitro evidence for a role of prostaglandin in gonadotropin secretion. ProstagZandin 1: 3-20, 1972. RASMUSSEN, H., M. PECHET, AND D. FAST. Effect of dibutyryl CAMP, theophylline, and other nucleotides on calcium and phosphate metabolism. J. CZin. Invest. 47: 1843-1850, 1968. RATLIFF, N. B., J. W. WILSON, E. MIKAT, D. HACKEL, AND T. C. GRAHAM. The lung in hemorrhagic shock. IV. The role of neutrophilic polymorphonuclear leukocytes. Am. J. PathoZ. 65: 325-332, 1971. RUSSELL, R. G. R., P. A. CASEY, AND H. FLEISCH. Stimulation of phosphate excretion by renal artery infusion of 3’,5’-AMP (cyclic AMP). A possible mechanism of action of parathyroid hormone. CaZcified Tissue Res. 2, Suppl: 54, 1968. SCORNIK, 0. A., AND A. C. PALADINI. Angiotensin blood levels in hemorrhagic hypotension and other related conditions. Am. J. Physiol. 206: 553-556, 1964. SELKURT, E. E. The renal circulation. In: Shock and Hypotension, Pathogenesis and Treatment, edited by C. Mills and J. H. Moyer. New York: Grune & Stratton, 1965, p. 157-169. SELKURT, E. E. Primate kidney function in hemorrhagic shock. Am. J. PhysioZ. 217: 955-961, 1969. SELKURT, E. E. The influence of aortic constriction on the positive free-water clearance of primate hemorrhagic shock. Proc. Sot. Exper. BioZ. Med. 140: 220-229, 1972. SELKURT, E. E. Role of ADH in the loss of renal concentrating ability in primate hemorrhagic shock. Proc. Sot. Exber. BioZ. Med. 142: 1310-1315, 1973. SELKURT, E. E. (editor). Pathological physiology of the cardiovascular system: physiology of shock. In: PhysioZogy. Boston: Little, Brown, 1971, 3rd ed. chapter. 18 (C), p. 409-424. SELKURT, E. E. Current status of renal circulation and related nephron function in hemorrhage and experimental shock. I. Vascular mechanisms; II. Neurohumoral and tubular mechanisms. Circulatory Shock, 1: 3-15, 89-97, 1974. SHIMIZU, K., T. KUROSAWA, T. MAEDA, AND Y. YOSHITSHI. Free water excretion and washout of renal medullary urea by PGE,. Japan. Heart J. 10: 437-455, 1969. SMITH, H. W., N. FINKELSTEIN, L. ALIMINOSA, B. CRAWFORD, AND M. GRABER. The renal clearance of substituted hippuric acid derivatives and other aromatic acids in dog and man. J. CZin. Invest. 24: 388-404, 1945. TERRAGNO, D. A., K. CROWSHAW, N. A. TERRAGNO, AND J. C. MCGIFF. Prostaglandin synthesis by bovine mesenteric arteries and veins. CircuZation Res. 36: Suppl. I, Hypertension, edited by N. M. Kapan 1-76-I-80, 1975. TERRA.GNO, N. A., A. J. LONIGRO, K. U. MALIK, AND J. C. McGIFF. The relationship of renal vasodilator action of bradykinin to the release of a prostaglandin E-like substance. Experientia 28: 437-439, 1972. VANDER, A. J. Control of renin release. Physiol. Rev. 47: 359-382, 1966. VANDER, A. J. Direct effects of prostaglandin on renal function and renin release in anesthetized dog. Am. J. Physiol. 214: 218221, 1968.

Downloaded from www.physiology.org/journal/ajplegacy at Tulane University (129.081.226.078) on February 15, 2019.