Hypertension and the kidney: Determinants of the response to antihypertensive therapy and their implications
Post on 18-Oct-2016
Hypertension and the kidney: Determinants of the response to antihypertensive therapy and their implications
Medicine has long recognized an association between hypertension and the kidney. The kidney may be a culprit or a victim in the process. As a culprit, the kidney may be responsible for the pathogenesis of hypertension in many patients, and in virtually all patients the renal response to antihypertensive therapy is a major determinant of its success or failure. In some patients, hypertension can lead to renal injury and even end-stage renal disease. Indeed, 25% of patients entering dialysis or transplant programs in the United States today have hypertension as the primary or sole mechanism, and another 25% have the complex combination of diabetes and hypertension as the cause. Antihypertensive therapy appears to be successful in preventing or arresting the renal response in accelerated hypertension, regardless of the treatment used to reduce blood pressure. However, treatment appears to be less successful in preventing the progression of moderate hypertension to end-stage renal disease. Substantial evidence suggests that angiotensin-converting enzyme inhibition and calcium channel blockade may prevent this progression when other antihypertensive therapy does not. The renal response to an angiotensin-converting enzyme inhibitor or a calcium channel-blocking agent appears to be determined by the pathogenetic features of the hypertension, and this may be an important determinant of the efficacy of the agents selected. Although still indistinct, the guidelines favoring selection of a specific antihypertensive agent are gradually emerging. (AM HEART J 1993;125:604-6.)
Norman K. Hollenberg, MD, PhD Boston, Mass.
Investigators who have approached the problem of the pathogenesis of essential hypertension from a wide variety of views have found it difficult to describe a possible pathogenetic sequence without implicating the kidney. l-6 In animal models7eg and in humans,lO* l1 studies based on transplantation have suggested that genetic information predisposing to hypertension is coded in the kidney. The pathogen- esis is likely complex in each patient, and probably differs from patient to patient. In some patients the hypertension is especially sensitive to sodium intake, whereas in others it is not. l2 In some patients plasma renin activity is elevated, and in others it is not5 Some patients have an increase in circulating cate- cholamines and presumably an increase in sympa- thetic nervous system activity; in others this evidence
From the Departments of Medicine and Radiology, Harvard Medical School and Brigham and Womens Hospital.
Reprint requests: Norman K. Hollenberg, MD, PhD, Brigham and Womens Hospital, 75 Francis St., Boston, MA 02115,.
Copyright 1993 by Mosby-Year Book, Inc. 0oor.L8703/93/$1.00 + .lO 4/o/42394
is absent. Unfortunately, no simple means of iden- tifying pathogenesis in the individual patient cur- rently exists, and few physicians attempt to do so when selecting therapy.
Whether the kidney is involved when hypertension is initiated, ample evidence suggests that the renal response to antihypertensive therapy is a crucial de- terminant of whether the antihypertensive therapy will be successful. The first recognition of this prin- ciple came from Guyton et al.* Their analog computer model of the cardiovascular system provided com- pelling evidence, which was strongly supported by observations from the clinic and the laboratory, that the kidney must be engaged to sustain hypertension however it is initiated. A second important principle was first clearly enunciated by Koch-Weser13 in his description of standard triple therapy, a combination of a vasodilator, a diuretic, and a @-adrenergic- blocking agent. A vasodilator is a rational first agent to select for the treatment of hypertension because resistance to blood flow is increased and a vasodila- tor reverses that fundamental hemodynamic abnor- mality. Koch-Weser also suggested that a vasodilator was ineffective when used alone, because it activated
Number 2. Part 2 Antihypertensive therapy and the kidney 605
several systems that limit the antihypertensive re- sponse. The addition of a /3-adrenergic-blocking agent and a diuretic could be seen as enhancing the therapeutic response primarily by blocking these re- active systems.
In this context, much of the recent advance in an- tihypertensive therapy makes a great deal of sense. Special attention has been given to angiotensin-con- verting enzyme (ACE) inhibition and calcium chan- nel blockade. Both approaches involve vasodilation, and both classes of agents have intrinsic to their ac- tion a special influence on the kidney, including pro- duction of a natriuresis and a reduction either in an- giotensin II formation or the vascular response to angiotensin II. The remarkable efficacy of these classes of agents probably reflects their substantial direct and indirect influences on the kidney.l
HYPERTENSION AND END-STAGE RENAL DISEASE
We have known since the 1960s that the overall morbid event rate is reduced by antihypertensive therapy in severe hypertension and that the explo- sive, rapidly progressive renal failure associated with accelerated hypertension can be prevented or ar- rested. During the past decade, however, a disturbing fact has emerged: The prevalence of end-stage renal disease in the United States has increased progres- sively, primarily as a result of an increase in the number of patients with renal failure resulting from hypertension or diabetic nephropathy., l6 Indeed, this has occurred despite a concurrent decrease in cardiovascular mortality from cerebrovascular events and even coronary events in this population and does not reflect malignant hypertension as a cause.
In one small study at a single center, Rostand et a1.17 found a substantial frequency with which other- wise unremarkable mild-to-moderate essential hy- pertension progressed to renal insufficiency. Al- though the criteria used to rule out primary renal disease were lax, this also applies to the data base from end-stage renal disease programs.15, l6 What emerged was that a category of patients, elderly, black, and male, was especially at risk,li which was supported by findings in the Hypertension Detection and Follow-up Program. s In this study there was a substantial frequency with which a significant de- cline in renal function occurred, again especially in black patients. Both studies used long-standing tra- ditional therapy, including a diuretic and methyl- dopa or a @-blocker as treatment. The possibility that the use of a diuretic contributed to the progressive loss of kidney function in selected patients is intrigu- ing, although clear evidence to support this is lacking. In an extended analysis of the European Working
Party Trial of High Blood Pressure in the Elderly, a study that was conducted primarily in whites and that used diuretics as treatment, a decline in renal function occurred in the treated group but not in the placebo group. It is difficult to argue with the authors conclusion that renal impairment is more likely to be a consequence of the diuretic treatment they used than of the hypertension per se. However, their experience in whites may not be reflected in other races. In the Baltimore Longitudinal Study on Aging, again conducted in a predominately white population, hypertension indeed accelerated the loss of kidney function associated with aging, and diuretic treatment may have also done ~0.~~
ANTIHYPERTENSIVE THERAPY AND PROGRESSION OF RENAL DESTRUCTION
By the late 1970s a consensus of informed opinion held that antihypertensive therapy prevented not only the renal injury of accelerated hypertension, but also could at least partially reverse the process once established. Moreover, the nephropathy of type I di- abetes mellitus could be delayed and slowed. Because these conclusions were straightforward and the indi- cation to treat hypertension was clear whether or not renal disease was present, essentially no controversy ensued. During the past decade a new debate has be- gun, which is based on several lines of investigation. The issue is currently not whether to treat hyperten- sion, all agree tha.t hypertension when pressed should be treated, but rather, do specific agents exert a dif- ferential influence on the renal response?
One major line of investigation that has influenced these thoughts has come from studies in animal models. A wide variety of insults can lead to chronic, progressive renal destruction. When a septic, immu- nologic, metabolic, or toxic injury is involved, the re- sponsible insult may or may not make a continuing contribution. The most unambiguous information has come from a model based on the remnant kid- ney. ;l If one kidney and a substantial portion of the other kidney are removed, a small residue of perfectly normal kidney tissue remains. In a manner identical to all forms of chronic progressive renal disease, that small remnant undergoes unremitting progressive glomerular sclerosis, which leads to a uremic death. The resultant end-stage kidney disease is essentially identical to that associated with other forms of injury.
The initial response of the remnant kidney can be seen as adaptive: With hypertrophy, blood flow to each glomerulus increases, as does its glomerular fil- tration rate. The result is short-term improvement in kidney function. However, as noted by Brenner,l
606 Hollenberg February 1993
Amermn Heart Journal
short-term adaptation might prove to be maladap- tive in the long run. The short-term increase in the glomerular filtration rate reflects an increase in renal plasma Now and a sharp increase in glomerular cap- illary hydrostatic pressure or glomerular capillary hypertension. By analogy, elevated hydrostatic blood pressure in any blood vessel can lead to local injury and destruction.
When protein intake was varied over a wide range, from 6, to 50% of total caloric intake, increased protein intake raised glomerular capillary pressure and increased injury. * Conversely, a reduction in protein intake reduced glomerular capillary pressure and plasma flow and retarded glomerular sclerosis. This accounts for recent attempts to use a very low- protein intake as part of the treatment of patients at risk of chronic, progressive renal destruction. Al- though hints of effectiveness have become available, translation int,o practical t,herapeutics has been lim- ited because a very low-protein intake is difficult for most patients to maintain.
Pharmacologic means were then used to explore the process. ACE inhibitors reduce not only arterial blood pressure but also glomerular capillary pres- sure. Sustained treatment with an ACE inhibitor in the remnant kidney model prevented glomerular sclerosis and progressive renal destruction. A recent review by Keane et al. cited 12 studies in a wide va- riety of models in the rat. All studies in which an ACE inhibitor was used demonstrated a clear retardation of progressive renal destruction, although a relatively limited range of models was used. The same review also described a series of studies in which some form of standard therapy, including a diuretic, a vasodila- tor, and an antiadrenergic agent, were combined to produce a similar decrease in blood pressure with use of a greater variety of models. In half of the studies, standard therapy appeared to be effective in reduc- ing renal injury; in the other half, it was not. In every study in which triple therapy and an ACE inhibitor were compared systematically, the ACE inhibitor was more effective.
Calcium channel-blocking agents have also been of substantial interest. The same review2 described six controlled studies in rats in which calcium channel- blocking agents had been used. Three showed unam- biguous efficacy in the prevention of progressive re- nal injury, but the other three did not. These early data probably underestimated the use of calcium channel blockers, because these agents may be difi- cult to use in rats, in which they are often ineffective in controlling hypertension, especially in some of t,he models selected for study. More recent observations in several models have suggested that calcium antag-
onists may have renal vascular effects and an influ- ence on renal destruction that varies wit,h the mod- el.,?- Although the predominant elect of calcium antagonists is generally on t,he afferent arteriole,x which would favor an increase in glomerular capillary pressure, in some models their use can actually lead to a decrease in glomerular capillary pressure.-
HYPERTENSION MODELS AND THE RENAL
VASCULAR RESPONSE TO CALCIUM CHANNEL BLOCKADE AND CONVERTING ENZYME INHIBITION
Prompted by these considerations, we undertook a systematic comparison of the state of the renal blood supply as a determinant of its response to either manidipine, a calcium antagonist, or enalaprilat, an ACE inhibitor.g Two models were studied: In one model, hypertension was induced by the long-term administration of desoxycorticosterone acetate (DOCA) along with a high-salt, diet, thereby produc- ing volume-dependent hypertension in the rat.RO,ll For the second model, we used the spontaneously hypertensive rats (SHRs), which are genetically de- termined to develop spontaneous hypertension. For this second model, substantial evidence suggests that participation of the renin-angiotensin system con- tributes to sustained renal vasoconstrict,ion.30-34 In contrast, the renin-angiotensin system is suppressed in the DOCA-salt. model.
As anticipated, enalaprilat and manidipine in- duced essentially identical depressor responses in the SHR but very different responses in the DOCA-salt model. Manidipine was equally effective in the DOCA-salt model and t.he SHR-Wistar-Kyoto (WKY), but enalaprilat was essentially ineffective in the volume-dependent model. Similarly, enalaprilat and manidipine induced very similar renal vascular responses in the SHR-WKY group but very different responses in the DOCA-salt model. Renal blood flow rise with manidipine was substantially greater in the DOCA-salt model than in SHRs and WKYs, whereas enalaprilat was essent.ially ineffective in the DOCA- salt group.
In SHRs. multiple lines of evidence suggest a renal contribution to the pathogenesis of hypertension. A particularly compelling experiment involves trans- plantation of the kidney, in which the blood pressure increase follows the kidney. In genetic backbreeding experiment,s. results point to a renal abnormali- ty. :1,i-37 The potentiated renal vascular response to an ACE inhibitor in SHRs enhances their renal vascu- lar reactivit~y to angiotensin II., x2 This constellation strikingly resembles the control of renal circulation in a group of patients with essential hypertension, the nonmodulators. In these essential hypertensive
Number 2, Part 2
patients, ACE inhibition substantially increases re- nal blood flow, enhances the renal vascular response to angiotensin II, and corrects the inability of the kidney to handle a sodium load.
The remarkable similarity of the depressor and re- nal vasodilator response to manidipine and enalapri- lat in SHRs may have reflected a relatively similar action in this strain. Manidipine strikingly reduced the renal vascular and pressor responses to angioten- sin II, as has been documented for calcium channel blockers in several studies.38-40
However, manidipine was as effective in reducing blood pressure in the DOCA-salt model as it had been in SHRs. Indeed, the influence of manidipine on the renal blood supply in this model was even more striking. The increase in renal blood flow was consid- erably greater with manidipine in the DOCA-salt model than in SHRs. The striking enhancement of vascular responses to the calcium antagonists may well have contributed to its undoubted efficacy in re- ducing blood pressure. Such a substantial renal vasodilator response must reflect an action on both afferent and efferent arteriolar resistance.23
The mechanism by which hypertension occurs in a volume-dependent, salt-sensitive model remains un- clear. One major line of investigation has involved the premise that a circulating digitalis-like factor con- tributes to hypertension in volume-dependent mod- els through an action on vascular smooth muscle so- dium/calcium exchange. 41,42 If manidipine influ- ences renal blood flow in the DOCA-salt model by blocking the local action of such a factor, the findings suggest a very striking action of this moiety on the renal blood supply. Certainly, substantial renal af- ferent arteriolar vasoconstriction has been demon- strated in this model.l
The available data suggest that not only the depressor response to a selected antihypertensive agent but also the renal vascular response will vary with the pathogenesis. In models of vasoconstriction with angiotensin as the mediator, the ACE inhibitor will have a preferential action. In contrast, in the volume-dependent models, calcium channel-block- ing agents may have a preferential renal action. There is preliminary but unambiguous evidence that the combination of a calcium antagonist and an ACE inhibitor may induce a renal vascular response that differs from either used alone.43
In light of evidence that the combination can be particularly effective in controlling difficult hyper- tension 44-47 there are compelling reasons to explore their emcacy when used jointly in the prevention of progressive renal injury. Clearly, the final chapter on the selection of an antihypertensive agent for the
Antihypertensiue therapy and the kidney 607
prevention of renal damage is not yet written. What is clear is that renal vascular tone and reactivity vary with different models of hypertension, and this is likely to have implications for treatment efficacy and selection.
Dahl LK. Metabolic aspects of hypertension. Ann Rev Med 1963;14:69-8. Guyton AC, Coleman TG, Cowley AW, Scheel KW, Manning RD, Norman RA. Arterial pressure regulation: overriding dominance of the kidneys in long-term regulation and in hy- pertension. Am J Med 1972;52:584-94. Dickinson CJ. Neurogenic hypertension with special reference to the central neurogenic effects of angiotensin. Circ Res 197O;(suppl 11)26:169-76. Ledingham JM. The etiology of hypertension: symposium on hypertension. Practitioner 1971;207:5-19. Laragh JH. Vasoconstriction-volume analysis for understand- ing and treating hypertension: the use of renin and aldoster- one profiles. Am J Med 1973;55:261-74. Tobian L. A viewpoint concerning the enigma of hypertension. Am J Med 1972;52:595-609. Bianchi G, Fox U, DiFrancesco GF, Giovanetti AM, Pagetti D. Blood pressure changes produced by kidney cross-transplan- tation between spontaneously hypertensive rats and nor- motensive rats. Clin Sci Mol Med 1974;47:435-48. Dahl LK, Heine M. Primary role of renal homografts in set- ting chronic blood pressure levels in rats. Circ Res 1975;36:692-6. Kawabe K, Watanabe TX, Sbiono K, Sokabe H. Influence on blood pressure of renal isografts between spontaneously hy- pertensive and normotensive rats utilizing the Fl hybrids. Jpn Heart J 1983;19:886-94. Curtis JJ, Luke RG, Dustan HP, Kashgarian M, Wheichel JD, Jones P, Diethelm AG. Remission of essential hypertension after renal transplantation. N Engl J Med 1983;309:1009- 15. Guidi E, Bianchi G, Rivolta E, Ponticelli C, Quart0 di Palo F, Minetti L, Polli E. Hypertension in man with a kidney trans- plant: role of familialversusother factors. Nephron 1985;41:14- 21. Fujita T, Henry WI, Bartter FC, Lake CR, Delea CS. Factors influencing blood pressure in salt-sensitive patients with hy- pertension. Am J Med 1980;69:334-44. Koch-Weser J. Vasodilator drugs in the treatment of hyper- tension. Arch Intern Med 1974;133:1017-27. Hollenberg NK. The kidney and effective antihypertensive therapy. Am J Cardiol 1985;56:52H-5H. National High Blood Pressure Education Program Workshop Group Report on Hypertension and Chronic Renal Failure. Bethesda. National Institutes of Health. National Heart, Lung, and Blood Institute, August 1990, NIH publication no. 90-3032. U.S. Renal data system. USRDS 1989 annual data report. Be- thesda: The National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, 1989. Rostand SG, Brown G, Kirk KA, Rutsky EA, Dustan HP. Re- nal insufficiency in treated essential hypertension. N Engl J Med 1989;321:692. Shulman NB, Ford CE, Hall WD, Blaufox D, Simon D, Lang- ford HG, Schneider KA. Proenostic value of serum creatinine and effect of treatment of hypertension on renal function: re- sults from the hypertension-detection and follow-up program. Hypertension 1989;(sunnl 13):180-92. DeLeeuw PW. Renal function in the elderly: results from the European working Party on High Blood Pressure in the Eld- erly trial. Am J Med 1991;9O(suppl 3A):45-9. Lindeman RD, Tobin JD, Shock NW. Association between
February 1993 Amermn Heart Journal
blood pressure and the rate of decline in renal function with age. Kidney Int 1984;26:861-8. Brenner BM. Nephron adaptation to renal injury or ablation. Am J Physiol 1985;249:467-70. Keane WF, Anderson S, Aurell M, de Zeeuw D, Narins RG. Povar G. Angiotensin converting enzyme inhibitors and pro- gressive renal insufficiency. Current experience and future di- rections. Ann Intern Med 1989;111(61:503-16. Anderson S. Renal hemodynamic effects of calcium antago- nists in rats with reduced renal mass. Hvpertension 1991: . 17:288-95. Dworkin LD. Effects of calcium channel blockers on experi- mental elomerular iniurv. J Am Sot Nenhrol 1990:1:S21-7. Dworkin LD, Levin RI,Benstein JA, Parker M, Dllian ME, Kim Y, Feiner HD. Effects of nifedipine and enalapril on glo- merular injury in rats with deoxycorticosterone-salt hyper- tension. Am J Physiol 1990;259:F598-604. Celsi G, Savin VJ, Henter JI, Sohtell M. The contribution ot ultrafiltration pressure for glomerular hyperfiltration in young nephrectomized rats. Acta Physiol Stand 1991;141:483-7. Nordlander M, Havu N. Effects of chronic felodipine treat- ment on renal function and mornhologv in SHR. Kidnev In- tern 1992;(suppl 36141:100-5. 1 ._ Loutzenhiser R, Epstein M. Effects of calcium channel antag- onists on renal hemodynamics. Am J Physiol1985;249:619-29. Hollenberg NK, Coletti C, Passan D. Hypertension, volume and vasoconstriction: studies on the renal blood supply in SHR, WKY, and DOCA-salt rat models. Hypertens Res 1992;15:3-11. Moreland RS. Lamb FS. Webb RC. Bohr DF. Functional ev- idence for increased sodium permeability in aortas from DOCA hypertensive rats. Hypertension Suppl 1984;6:188-94. Iversen BM, Morkrid L, Ofstad J. Afferent arteriolar diame- ter in DOCA-salt and two-kidney one-clip hypertensive rats. Am .J Physiol 1983;14:F755-62. Guidi E, Hollenberg NK. Differential pressor and renal vascular reactivity in AI1 in spontaneously hypertensive and Wistar-Kyoto rats. Hypertension 1987;9:591-7. Arendshorst WJ. Chatziantoniou C. Daniels FH. Role of an- giotensin in the renal vasoconstriction observed during the development of genetic hypert,ension. Kidney Intern 1990: X):9%-6. Matsushima Y, Kawamura M, Atabane S, Imanishi M. Kuramochi M, Ito K. Omae T. Increases in renal AI1 content
and tubular AI1 receptors in prehypertensive sp~,ntann,us~~ hypertensive rats. .7 Hyperiens 1988;6:791-6 Kawabe K. Watanabe TX. Shiono K, Sokabe H. Influence on blood pressure or renal isoprafts between spontaneously- hypertensive and normotensive rats, utilizing 1 he FI I) h? brids. Jpn Heart J 1978;20:886-94. Harrap SB. Genetic analysis (11 blood pressure and sodium balance in spontaneously hypertensive rats. Hypertension 1986;8:57%83. Harrap SB, Doyle AE. Genetic co-segregation ofrenal hemcr- dynamics and blood pressure in the spontaneously hyperten- sive rat. Clin Sci 1988:74:63-Y. Blackshear .JL. Garnic D, Williams GH, Harrington DP, Hol- lenberg NK. Exaggerated renal vascular response to calcium entry blockade in first degree relatives of essential hyperten- sives: possible role of intrarenal AII. Hypertension 1987; 9:384-g. Loutzenhiser R, Epstein M. Effect.s of calcium antagonists on renal hemodynamics. Am d Physiol 1985;249-F619-29. Goldberg .JP. Schrier RW. Effect of calcium membrane block- ers on in vivo vasoconst,rictor properties of norepinephrine, AII, and vasopressin. Miner Electrolyte Metab 1984;10:178- 8.?. Haddy FJ, Overbeck HW. The role of humoral agents in vol- ume expanded hypertension. Life Sci 1976;19:935-48. Blaustein MP, Hamlyn ,JM. Role of a natriuretic factor in es- sential hypertension: an hypothesis. Ann Intern Med 1983;9A(prt 21:785-g. Carmines PKI, Navar LG. Disparate effects of Ca channel blockade on afferent and efferent arteriolar responses to ANG II. Am .I Physiol 1989;256:FlO15-20. Mimran AM, Ristein J. Effect ofchronic nifedipine in patients inadequately controlled by a converting enzyme inhibitor and a diuretic. .I Cardiovasc Pharmacol 1985:7:S92-5. Brouwer RMJ , Bolli P. Erne P. Conen D, Kiowski W. Buhler FR. Antihypertensive treatment using calcium antagonists in combination with captopril rather than diuret its. ,J Cardiovasc Pharmacol 1985;7:88-91. Singer DRJ. Markandu ND, Shore AC, MacGregor GA. Cap- t.opril and nifedipine in combination for moderate to severe essential hypertension. Hypertension 1987;9:629-33. Pieri R. Nardecchia A, Pirrelli A. Combined nifedinine and captopril treatment in moderately severe primary hyperten- sion. Am ,I Nephrol 1986:6(suppl l):lll-4.