Pharmacodynamic Mechanisms for Diuretic Antagonism

Since the clinically important diuretics work from the luminal side of the renal tubule, pharmacodynamic interactions between NSAIDs and diuretics can be

FIGURE 2. Dose-response curve depicting the relationship between urinary sodium excretion and urinary furosemide excretion rates in volunteers before and after indomethacin administration. After indomethacin treatment subjects required greater rates of furosemide excretion to achieve any desired rate of sodium excretion. (From Chennavasin P., Seiwell, R., Bater D, (1980) Pharmacoki-netic-dynamic analysis of the indomethacin-furosemide interaction in man. J. Pharm. Exp. Ther. 215; Fig. 2, p. 66. with permission).

Urinary Furosemide Excretion Rate |ig / min

FIGURE 2. Dose-response curve depicting the relationship between urinary sodium excretion and urinary furosemide excretion rates in volunteers before and after indomethacin administration. After indomethacin treatment subjects required greater rates of furosemide excretion to achieve any desired rate of sodium excretion. (From Chennavasin P., Seiwell, R., Bater D, (1980) Pharmacoki-netic-dynamic analysis of the indomethacin-furosemide interaction in man. J. Pharm. Exp. Ther. 215; Fig. 2, p. 66. with permission).

determined by examination of the relationship between urinary diuretic concentration and urinary sodium excretion. Pharmacodynamic interactions shift this dose-response curve to the right so that in the presence of the interaction a higher concentration of diuretic is required in the urine to achieve any given level of sodium excretion. The effect of indomethacin on the relationship between urinary furosemide concentration and urinary sodium excretion is shown in Fig. 2. Qualitatively similar curve shifts have been reported for the effect of salicyates, ibuprofen, naproxen, or sulindac on furosemide induced natriuresis and for effects of indomethacin on bumetanide and hydrochlorothiazide response in man. Curve shifts have also been reported for effects of indomethacin or meclofenamate on furosemide, acetazolamide, amiloride, and bumetanide natriuresis in experimental animals.

Multiple pharmacodynamic mechanisms could account for the ability of NSAIDs to decrease diuretic responses. Recognition that these are based on inhibition of prostaglandin synthesis does not allow identification of a specific mechanism because of the variety of prostaglandin mediated processes present in the kidney [1], Additionally, administration of NSAIDs rarely inhibit renal prostaglandin synthesis by greater than 80%. Thus, it is possible that the amount of prostaglandin production remaining may be sufficient to support prostaglandin dependent processes in some renal locations but not others. This makes assessment of the contribution of specific pharmacodynamic mechanisms to the overall antagonism of diuretic response difficult to quantify. NSAIDs could induce pharmacodynamic antagonism of diuretic response through two general categories of mechanisms: hemodynamic mechanisms and direct antagonism of the diuretic's tubule effect. These mechanisms need not be mutually exclusive.

Hemodynamic Mechanisms

Hemodynamic mechanisms potentially important in the attenuation of diuretic response by NSAIDs include reductions in glomerular filtration rate (GFR) and alterations in total renal blood flow or in intrarenal blood flow distribution. The first limits solute delivery to the tubule, while the second alters peritubular physical factors to favor sodium reabsorption. Reductions in GFR and, thus, reductions in filtered sodium load have been inconsistently reported in NSAID treated animals and humans during administration of loop and thiazide diuretics. Because of the magnitude of the daily filtered sodium load (approximately 20,000 mEq/day), a reduction in GFR undoubtedly plays a major role in attenuating the natriuretic response to diuretics in circumstances where it occurs. Although decreases in GFR are inconsistently observed in normal individuals during NSAID administration, reductions in GFR following prostaglandin synthesis inhibition are not infrequent in conditions associated with a decrease in effective extracellular fluid volume or an increase in plasma renin activity such as congestive heart failure, nephrotic syndrome, and cirrhosis. While this mechanism has not been considered a significant contributor to diuretic antagonism in most experimental studies, it is likely to play an important role in clinical settings where NSAIDs and diuretics are used. Even in these situations the contribution of this mechanism may not be fully appreciated because of the insensitivity of the techniques employed to estimate GFR.

Reductions in renal blood flow or changes in intrarenal blood flow distribution were some of the first mechanisms proposed to explain how NSAIDs alter diuretic action. The attractiveness of this hypothesis is enhanced by observations that furosemide increases urinary prostaglandin E2 excretion by increasing availability of arachidonic acid, by decreasing prostaglandin degredative enzymes, and at high doses by directly stimulating prostaglandin E2 synthesis. Furosemide stimulation of prostaglandin release induces vasodilation and increases venous capacitance. These events have been proposed to account for the observation that furosemide administered intravenously to patients in left heart failure produces a fall in pulmonary capillary wedge pressure prior to producing a significant natriuresis. The finding that this response is blocked by NSAIDs and does not occur in nephrectomized subjects suggests that these events are mediated by furosemide stimulation of renal prostaglandin release. Stimulation of urinary prostaglandin excretion has also been reported following oral administration of loop, thiazide, and potassium-sparing diuretics in some human studies [10]. The intravenous administration of furosemide to anesthetized dogs increases total renal blood flow and redistributes this flow to the midcortical and away from juxtamedullary areas of the kidney [4, 12]. An increase in juxtamedullary flow has also been observed in some studies [4]. Indomethacin reduces basal total renal blood flow and blood flow to juxtamedullary areas and prevents the increase in blood flow induced by furosemide [4]. Increases in total renal blood flow have also been reported following intravenous administration of ethacrynic acid and bumetanide. Indomethacin blocks the increases in renal blood flow induced by these agents as well. Whether antagonism of the hemodynamic changes induced by loop diuretics is the primary mechanism accounting for the antinatriuretic effect of NSAIDs is not clear. In support of this hypothesis Nies and associates have found that indomethacin antagonized furosemide natriuresis only under conditions where it also antagonizes furosemide induced increases in renal blood flow [11]. On the other hand, attenuation of furosemide natriuresis by indomethacin and meclofena-mate can occur in the absence of alterations in renal blood flow [5, 7]. NSAID induced reductions in renal blood flow have been reported in the absence of alterations in the natriuretic response to loop diuretics. Assessment of these studies is complicated because changes in intrarenal blood flow, especially in the medullary circulation, may not be reflected by changes in total renal blood flow. Indeed, there is evidence that blood flow in the medulla may be regulated independently of total renal blood flow largely due to the local influences of prostaglandins, angiotensin II, and perhaps nitric oxide. We have measured furosemide response in indomethacin treated rats in which the reduction in medullary plasma flow usually associated with NSAID administration was prevented by blocking angiotensin II mediated vasoconstriction [7]. Consistent with reports from other investigators the attenuated furosemide natriuresis found in indomethacin treated rats was associated with a significant reduction in medullary plasma flow. However, preventing the fall in medullary plasma flow associated with indomethacin administration did not alter indomethacin capacity to antagonize furosemide natruresis. In aggregate the available studies suggest that changes in renal hemodynamics are not a prerequisite for antagonism of loop diuretics by NSAIDs.

Tubular Mechanisms

Prostaglandin synthesis inhibition could also blunt natriuretic responses to diuretics by direct effects on tubule reabsorption. This could occur in the tubule segment directly inhibited by the diuretic or in tubule segments proximal or distal to the diuretic's location of action. The magnitude of the reduction in sodium excretion observed during NSA1D antagonism of loop and thiazide diuretics would favor more proximal tubule segments as the location for this event.

Whether prostaglandins play a direct role in the regulation of salt and water reabsorption in the anatomical proximal tubule is controversial. It is equally unclear whether inhibition of prostaglandin synthesis has direct effects on transport at this location. In micropuncture studies examining potential tubular locations for NSAID antagonism of diuretic response, we have shown that neither indomethacin nor meclofenamate alters chloride or fluid delivery to the late proximal convoluted tubule in rats receiving furosemide (Fig. 3). We have reported similar findings during indomethacin antagonism of hydrochlorothiazide natriuresis [8]. These findings are consistent with the observation of others that inhibition of prostaglandin synthesis does not alter proximal tubule function in the rat. However, micropuncture methodology examines only 60% of the proximal tubule. In humans there is also little objective evidence that major changes in proximal tubule reabsorption occur during antagonism of diuretic

FIGURE 3. Fractional chloride reabsorption in proximal, loop, and distal tubule segments during furosemide natriuresis in control rats (open bars), indomethacin treated rats (hatched bars), and meclofenamate treated rats (solid bars). Indomethacin and meclofenamate had no effect on proximal tubule reabsorption but significantly blunted furosemide inhibition of loop reabsorption. Reabsorption in the distal tubule was lower in indomethacin and meclofenamate treated rats. * P<.05 vs control at same tubule location. From Kirchner (1985, Fig. 1, p. F701.) with permission.

Proximal Tubule Loop Distal Tubule

FIGURE 3. Fractional chloride reabsorption in proximal, loop, and distal tubule segments during furosemide natriuresis in control rats (open bars), indomethacin treated rats (hatched bars), and meclofenamate treated rats (solid bars). Indomethacin and meclofenamate had no effect on proximal tubule reabsorption but significantly blunted furosemide inhibition of loop reabsorption. Reabsorption in the distal tubule was lower in indomethacin and meclofenamate treated rats. * P<.05 vs control at same tubule location. From Kirchner (1985, Fig. 1, p. F701.) with permission.

effects by NSAIDs. Some human studies on effects of prostaglandin synthesis inhibition on renal function have shown altered free water clearance, a finding interpreted as evidence of a role for prostaglandins in proximal tubule reabsorption. Interpretation of free water clearance data during inhibition of prostaglandin synthesis is complicated, however, by independent effects of prostaglandin inhibition on collecting duct water uptake. Thus, even though increased solute reabsorption in this location would be an attractive mechanism for explaining the finding that NSAIDs antagonize multiple diuretic agents, whether NSAIDs blunt diuretic responses through alterations in sodium and water transport in the proximal tubule remains unknown.

There is convincing evidence that the loop segment is an important location for the attenuation of diuretic response during prostaglandin synthesis inhibition. Micropuncture studies have shown that rats receiving intravenous furosemide and indomethacin or meclofenamate have twice the fractional and absolute chloride, and thus presumably sodium, reabsorption in the loop segment as do rats receiving furosemide alone (Fig. 3). We have also shown that indomethacin treated rats had 40% less inhibition of chloride reabsorption than control rats when furosemide was microperfused directly into isolated loop segments [6]. Furthermore, the inhibitory effect of indomethacin could be abolished by the systemic or intra tubular administration of prostaglandin E2. These observations categorically demonstrate that antagonism between NSAID and loop diuretics can result from direct antagonism of furosemide's effect on tubule transport and can occur independently of furosemide or NSAID induced alterations in renal blood flow or in renal blood flow distribution. How NSAIDs antagonize loop diuretics at this location is as yet undetermined. Indomethacin does not alter furosemide induced changes in short circuit current in frog skin, suggesting that indomethacin effect is not the result of interference with furosemide binding to the Na-K-2Cl cotransporter or to the transport process itself. We have shown that indomethacin does not blunt furosemide response in ADH deficient Brattleboro rats despite equivalent stimulation of urinary prostaglandin excretion by furosemide and suppression by indomethacin. Indomethacin antagonism of furosemide response could be restored by ADH replacement. In the rat ADH stimulates loop sodium chloride transport and this effect is inhibited by prostaglandin E2. Thus NSAIDs may increase loop uptake by interfering with prostaglandin modulation of ADH stimulated sodium chloride transport. Whether this mechanism is important in diuretic antagonism in humans is unclear as ADH is not felt to modulate loop transport in man. The loop segment may be the location for NSAID induced antagonism of other classes of diuretics as well. We have shown that the loop segment is the only nephron segment demonstrating increased reabsorption during indomethacirfs antagonism of hydrochlorothiazide response [8]. Brater has proposed that the antagonism between NSAIDs and potassium-sparing diuretics occurs as a result of increased sodium uptake in the loop as well [2]. Whether increased sodium uptake in nephron segments beyond the loop of Henle contributes to diuretic antagonism by NSAIDs has not been extensively examined. Chloride reabsorption in the distal convoluted tubule was found to be suppressed by NSAIDs during furosemide administration and unchanged during hydrochlorothiazide administration in the rat [5, 8]. Others have reported that all of the reduction in sodium excretion in meclofenamate treated rats and two-thirds of the reduction in sodium excretion in indomethacin treated rats observed during furosemide administration occurred before the inner medullary collecting duct [13]. Thus most experimental evidence implicates the loop segment as the primary location for the diuretic antagonism observed during NSAID administration.

Another potential tubular mechanism for antagonism between NSAIDs and spironolactone is competition for mineralocorticoid receptors. Displacement studies have shown that the sequence of binding affinity for these receptors is aldosterone —■> spironolactone —> phenylbutazone —» aspirin —> indomethacin. The concentration of indomethacin required to displace aldosterone from its receptor is greater than would occur clinically but aspirin and phenylbutazone concentrations are within concentrations which could occur clinically. The extent to which this mechanism contributes to antagonism of potassium-sparing diuretics is unclear.

Blood Pressure Health

Blood Pressure Health

Your heart pumps blood throughout your body using a network of tubing called arteries and capillaries which return the blood back to your heart via your veins. Blood pressure is the force of the blood pushing against the walls of your arteries as your heart beats.Learn more...

Get My Free Ebook


Responses

  • catherine bennett
    Is a nsaid and diuretic antagonistic?
    6 years ago

Post a comment