Glomerular hemodynamics and the renin-angiotensin system in patients with type 1 diabetes mellitus

Transcription

1 Kidney International, Vol. 63 (2003), pp VASCULAR BIOLOGY HEMODYNAMICS HYPERTENSION Glomerular hemodynamics and the renin-angiotensin system in patients with type 1 diabetes mellitus NORMAN K. HOLLENBERG, DEBORAH A. PRICE, NAOMI D.L. FISHER, M. CECILIA LANSANG, BRUCE PERKINS, MICHAEL S. GORDON, GORDON H. WILLIAMS, and LORI M.B. LAFFEL Departments of Medicine and Radiology, Brigham and Women s Hospital and Harvard Medical School, Boston, Massachusetts, USA Glomerular hemodynamics and the renin-angiotensin system sion, and the beneficial action of angiotensin-convertingin patients with type 1 diabetes mellitus. enzyme (ACE) inhibition on the kidney involves prefer- Background. Many studies have reported that blocking the ential efferent arteriolar dilation [1 6]. A related elerenin-angiotensin-system (RAS) with an angiotensin-convertment is the suggestion that the effects of ACE inhibition ing enzyme (ACE) inhibitor or an angiotensin receptor blocker in the patient with diabetes mellitus leads to an increase in on kinin metabolism also contribute to efferent arteriolar renal plasma flow (RPF), no change in glomerular filtration dilatation as a major part of therapeutic efficacy [7 11]. rate (GFR), and a fall in filtration fraction. This constellation is As the construct is simple and easy to understand, generally attributed to predominant efferent arteriolar dilation. logical and internally consistent, and supported by sub- Methods. This study examined the renal hemodynamic restantial data, it has been subject to little thoughtful critisponse to blocking the RAS with both captopril and candesartan on separate days in 31 patients with type 1 diabetes mellitus. cism. We were surprised, therefore, by the fact that the Results. There was a wide range of changes in RPF and angiotensin II subtype 1 (AT 1 ) receptor blocker, irbesar- GFR in response to the two agents, each administered at the tan, not only increased renal plasma flow (RPF) in patop of its dose-response range. The RPF response to the two tients with type 2 diabetes mellitus and nephropathy, it agents was strongly concordant (r 0.65; P 0.001), as was the GFR response (r 0.81; P 0.001). Moreover, there was also increased glomerular filtration rate (GFR) substan- a strong correlation between the RPF response and the change tially in some patients [12]. Indeed, experts in the field in GFR with each agent (r 0.83 and 0.66; P 0.01). A expressed concern that an increase in GFR might limit significant rise in RPF was followed by a rise in GFR. The the therapeutic efficacy of AT 1 receptor blockers in such RPF dependency of GFR in the type 1 diabetics suggests patients. We now know that AT 1 receptor blockers are strongly that glomerular filtration equilibrium exists in the gloeffective in decreasing progression of ESRD [13 15], meruli of the diabetic kidney: Simple notions of local control based on afferent:efferent arteriolar resistance ratios are too especially so early in the course [13] when the rise in simplistic. GFR was most striking in our studies [12]. There are Conclusion. Our data suggest that the intrarenal RAS is alternatives to afferent and efferent-resistant ratios in activated in over 80% of patients with type 1 diabetes mellitus. the control of GFR [16, 17], the subject of this study. Abundant evidence suggests that this activation predisposes to diabetic nephropathy. Measurements of the renal clearance of markers of RPF and GFR the only approach currently available for their assessment in the intact human have long been A construct explaining glomerular hemodynamics in believed to enjoy a rather limited signal-to-noise ratio the kidney at risk of progression to end-stage renal disintra-individual concordance of changes in para-amino- [18]. We were surprised, therefore, by the remarkable ease (ESRD) is so widely held today that it seems to hold the status of received truth. According to this concept a hippurate (PAH) clearance in response to an ACE inhib- preferential action of angiotensin II (Ang II) on the itor and a renin inhibitor in healthy humans [19], and to efferent arteriole contributes to glomerular hypertenwith type 1 diabetes mellitus [20]. Moreover, the rise in an AT 1 receptor blocker and ACE inhibitor in patients RPF in some patients with diabetes mellitus was sufficient [12, 20] to influence GFR if RPF dependency of Key words: ACE-I, angiotensin receptor blockade, intrarenal renin system, filtration equilibrium. GFR existed in that setting. Thus, we undertook this study with two goals in mind: (1), to examine the RPF depen- Received for publication February 21, 2002 and in revised form May 14, 2002 dency of GFR in type 1 diabetes mellitus; and (2), to Accepted for publication August 5, 2002 exploit studies in which the response to an ACE inhibitor, captopril, and an AT 1 receptor blocker, 2003 by the International Society of Nephrology candesartan, 172

2 Hollenberg et al: Glomerular dynamics 173 was compared systematically to provide an index of the where balance was achieved on a controlled diet. All reproducibility of the clearance measures, and to ascertain subjects were placed on a high-salt isocaloric diet starting the mechanism of the ACE inhibitor-induced re- two days prior to admission and continuing throughout sponses. Specifically of interest was the question of the hospitalization, with a daily sodium intake of 200 mmol. whether the reduction of Ang II formation induced by Daily dietary potassium (100 mmol) and fluid intake the ACE inhibitor could account for this effect. (2500 ml) were constant. Twenty-four-hour urine sam- ples were collected daily and analyzed for sodium, potassium, creatinine and protein. Twenty-four-hour urine so- METHODS dium and potassium content was expected from the diet Subjects and protocols (Table 1), as were PRA and plasma aldosterone concentrations. A total of 31 men and women with type 1 diabetes The protocol was approved by the Human Sub- mellitus were studied who were admitted sequentially jects Committee of the institution, and written informed to this study. They ranged in age from 17 to 62 years consent was obtained from each subject. (mean SEM; ). The duration of diabetes ranged from 2 to 40 years (mean ). Type 1 diabetes Renal hemodynamic and hormonal responses to the mellitus was diagnosed according to accepted guidelines ACE inhibitor, captopril and the AT 1 receptor [21]. Body weight ranged from 55 to 110 kg with an blocker, candesartan average of kg. Body mass index ranged from Each subject participated in two experimental days to 35.6 kg/m 2 with an average of kg/m 2. On the morning of each study day an intravenous cathe- Body mass index exceeded 28 in four of the patients. ter was placed in each arm of each subject, one for Blood pressure averaged / mm Hg, infusion of PAH, inulin, and dextrose 5% in water and and exceeded a systolic BP of 135 mm Hg in four. Dia- the other for blood sampling. A third intravenous line stolic BP did not exceed 85 mm Hg in any patient. Hemoglobin was placed for continuous infusion of insulin that was A 1C (HbA 1c ) ranged from 4.8 to 10.7%, and aver- started at units/kg/h. Blood glucose was measured aged %. Hemoglobin A 1c exceeded 9% in four every 30 minutes (Precision PCX; Abbott Laboratories, subjects. Similarly, fasting blood sugar averaged 144 Chicago, IL, USA). The insulin infusion was adjusted to 8.8 mg/dl, exceeding 200 mg/dl in three. Serum creatinine maintain blood glucose below renal threshold but withone concentration was less than 1.4 mg/dl in all but out inducing hypoglycemia, at levels of about 100 to 140 patient in whom serum creatinine was 2.0 mg/dl. mg/dl. The subjects were supine and had been fasting All were otherwise healthy, free of sustained proteinuria for at least eight hours. Each study day began with a 60- and other complications of diabetes. Three 24-hour urine minute baseline infusion of PAH and inulin prior to drug collections were made in search of microalbuminuria. administration, to determine baseline renal plasma flow We identified microalbuminuria, reflecting a value exceeding (RPF) and glomerular filtration rate (GFR) respectively. 17 g/mg creatinine in males, and 25 g/mg Hormonal measurements were made on blood samples creatinine in females in five subjects. In three of the five, obtained at baseline and four hours after drug adminis- all three urine collections showed microalbuminuria. In tration while the subjects were lying supine. Peak RPF the other two subjects, two of the three collections showed was identified as the two largest sequential measures of an increase. There was no difference in the renal vascular PAH clearance. The term peak GFR designates GFR response to pharmacological interruption of the reninangiotensin-system measurements made at the same time as peak RPF. in those with microalbuminuria, and The study was designed to compare the renal hemody- those who lacked it. There were five patients in whom namic response to captopril and to candesartan. On the hypertension had been diagnosed. In each, treatment first morning the patients received captopril, 25 mg PO. involved an ACE inhibitor or an angiotensin antagonist, On the next morning, the patients received candesartan, and in three additional patients who were free of hyper- 16 mg PO. These doses were chosen because both represent tension, ACE inhibition had been undertaken as a prophylactic. the top of the relationship between dose and RPF ACE inhibitors and angiotensin antagonists response. were discontinued 10 days before the renal hemodynamic Blood pressure was recorded during each infusion by study. Our study plan included the use of alpha methyl- an automatic recording device (Dinamap; Critikon, Norderstedt, dopa to control hypertension should therapy be needed, Germany) at five-minute intervals. but in no case was it required. There was no difference in the renal vascular response to blocking the renin system Renal clearance studies in those with or without hypertension or prior ACE Para-aminohippurate (Merck, Sharp & Dohme, Rah- inhibitor use. The subjects were studied during admission way, NJ, USA) and inulin (Inutest; Fresenius Pharma to a metabolic ward at the General Clinical Research Austria GmbH, Linz, Austria) clearances were assessed Center (GCRC) at the Brigham and Women s Hospital after metabolic balance was achieved. A control blood

3 174 Hollenberg et al: Glomerular dynamics Table 1. Baseline characteristics of 31 subjects in high sodium balance Table 2. Baseline renal hemodynamics of 31 subjects in high sodium balance Parameters Mean SEM Captopril vs. Age years candesartan Sex distribution male/female 13/18 Captopril Candesartan P value Race All Caucasian Body mass index kg/m Renal plasma flow Systolic blood pressure mm Hg ml/min/1.73 m 2 Diastolic blood pressure mm Hg Baseline Duration of diabetes years Glomerular filtration rate Hemoglobin A1C % ml/min/1.73 m 2 Fasting blood sugar mg/dl Baseline Serum creatinine mg/dl Serum sodium meq/l Serum potassium meq/l hour urine sodium meq hour urine potassium meq 59 4 vascular and PRA responses to captopril and candesar- Plasma renin activity ng/ang I/mL/h Plasma aldosterone ng/dl tan. The subjects were then divided into two groups, those in whom the RPF response to captopril was normal and those in whom the response was accentuated, to determine if there were differences in renin activity. Fisher s sample was drawn, and then loading doses of PAH (8 exact test, t test and Mann-Whitney rank sum test mg/kg) and inulin (50 mg/kg) were given intravenously. were used to compare the characteristics and the renal A constant infusion of PAH and inulin was initiated hemodynamic and PRA responses of these two groups. immediately at a rate of 12 mg/min and 30 mg/min, reaccount for possible confounding effects of baseline char- Analysis of covariance (ANCOVA) was performed to spectively, with an IMED pump (IMED Corp., San Diego, CA, USA). This achieved a plasma PAH concentration acteristics on RPF responses on the two study days. in the middle of the range in which tubular secre- tion dominates excretion. At this plasma level of PAH, RESULTS clearance is independent of plasma concentration, and Baseline characteristics of the subjects are listed in represents about 90% of RPF when corrected for indi- Table 1. The high salt balance, as evidenced by a 24- vidual body surface area. Likewise, at the level of plasma hour urine sodium excretion of meq, resulted inulin achieved, inulin clearance reflects GFR. RPF and in suppression of PRA at baseline, as anticipated. GFR determinations were made at baseline and at 45- Renal plasma flow at baseline was not significantly minute intervals thereafter until 225 minutes ( 4 h) different for captopril and candesartan (Table 2). Both while the subjects were supine. captopril and candesartan caused a significant rise in Laboratory procedures RPF. Following captopril RPF rose by ml/min/ 1.73 m 2 (P 0.01). Renal plasma flow rose by more than Blood samples were collected on ice and spun immedi- 20 ml/min/1.73 m 2 the upper limit of the response in ately, and the plasma was frozen until assay. Urinary so- normal subjects of the renal vascular response to captodium and serum potassium levels were measured using the pril on a high salt diet in all but 6 of the 31 subjects. ion-selective electrode (ISE). PAH and inulin were mea- Candesartan likewise induced a substantial rise in RPF sured using an autoanalyzer technique. Plasma renin ac- of ml/min/1.73 m 2 (P 0.001). The same six tivity (PRA) and aldosterone were determined by radio- subjects showed limited responses to candesartan. The immunoassay [22, 23]. Hemoglobin A1C was measured response to captopril and candesartan were highly correby high performance liquid chromatography (HPLC). The lated (Fig. 1; r 0.65; P 0.001). normal range is 4.4 to 6.3%. Baseline and peak GFR values were likewise not significantly Statistical analyses different between the two drugs. There was no significant change in GFR, on average, either in response Group means were calculated with the standard error to captopril or to candesartan (P 0.74 and P 0.45, of the mean (SEM) as the index of dispersion. For renal respectively). By analogy with the RPF response to caphemodynamics data, the baseline value taken was the topril and to candesartan, GFR responses to both agents average of three pre-drug determinations, and the peak were also highly correlated (Fig. 2; r 0.81; P 0.001). response was the average of the two highest consecutive Individuals in whom captopril seemed to raise GFR also values. showed a rise in GFR with candesartan. The individuals Pearson s correlation was used to test the association in whom captopril appeared to drop GFR also showed of the renal response to candesartan with the response a drop with candesartan. to captopril. Paired t test was used to compare the renal There was a clear correlation between the renal vaso-

4 Hollenberg et al: Glomerular dynamics 175 Fig. 1. Renal plasma flow of captopril versus candesartan. Relationship between the renal hemodynamic response to angiotensin-converting enzyme (ACE) inhibition by captopril, and to angiotensin II type 1 (AT 1 ) receptor blockade by candesartan in 31 patients with type 1 diabetes mellitus. Note the concordance between the renal hemodynamic response to captopril and to candesartan (r 0.65; P 0.001; y 0.75x 28.90; f 20.9). Fig. 2. Glomerular filtration rate of captopril versus candesartan. Relationship between changes in glomerular filtration rate (GFR) induced by candesartan and by captopril in the same patients as in Figure 1. Note that the responses to the two agents show close accord (r 0.81; P 0.001; y 0.91x 2.75; f 48.3; N 28). When GFR rose in response to captopril, it also rose in response to candesartan. When GFR fell in response to captopril, it also fell in response to candesartan. dilator response to candesartan, indicated by a change in RPF, and the change in GFR (Fig. 3A; r 0.83; P 0.001). The individuals in whom GFR rose were largely those in whom candesartan induced a substantial rise in RPF. Similar relationships were evident between the renal hemodynamic response to captopril and the influence of captopril on GFR (Fig. 3B; r 0.66; P 0.002). Although it was not the primary focus of this study, it was evident that the six individuals who failed to show an enhanced response to blockade of the RAS with cap- Fig. 3. PAH versus inulin. (A ) Relationship between renal plasma flow (RPF) response to candesartan as measured by PAH clearance, and simultaneous change in glomerular filtration rate (GFR) as measured by inulin clearance. Note the large increases in RPF were associated with a substantial increase in GFR (r 0.83; P 0.01), evidence of the RPF dependency of GFR in this situation. (B ) Relationship between RPF response to captopril in the same patients, and its influence on GFR. Note the similar relationship between RPF and GFR following captopril (r 0.66; P 0.01): GFR changes are RPF-dependent. topril and candesartan differed from the others in several possibly important ways. They were older (38 6 vs years), had a longer duration of diabetes mellitus (27 5 vs years), and despite the increased duration were still free of proteinuria and hypertension. They did not differ from the others in the indices of glycemic control. DISCUSSION On average, both captopril and candesartan induced a significant increase in RPF in patients with type 1 diabetes mellitus. Again, on average GFR neither rose nor fell, as reported in many studies [2 6]. The strong concordance between the RPF response induced by captopril and by candesartan indicates that the rise in RPF in individual

5 176 Hollenberg et al: Glomerular dynamics effect on GFR [18, 31, 32]. This favors glomerular filtration disequilibrium in normal humans. With the develop- ment of diabetes mellitus, there is a substantial increase in both renal and glomerular size, with concomitant changes in the glomerular capillaries. The findings in this study raise the interesting possibility that one physiologi- cal consequence of this morphologic change is the development of glomerular filtration equilibrium, and thus RPF dependency of GFR. The fact that diabetes mellitus is a volume expanded state, also would favor filtration equilibrium. There are alternatives to filtration equilibrium as an explanation for the rise in GFR. Glomerular intermittency, which was corrected by ACE inhibition and angiotensin receptor blockade, would provide an alternative explanation. There is little evidence of glomerular intermittency in species beyond the amphibian [18], and intermittency would be especially unlikely in view of the very high baseline GFR. Why have these relationships been missed in earlier studies? In general, the renal hemodynamic response to ACE inhibition or angiotensin receptor blockers in the patients with renal disease or at risk of renal injury has been presented as group means. When examined in this way, we found in our current study what others have found and reported in the past: RPF rose, GFR on average did not change, and as a consequence filtration frac- tion fell. Over the past decade, however, we have come to appreciate that the clearance method, performed care- fully, is sufficiently precise that it can be applied to the individual measure. In a double blind comparison in healthy humans we found a striking concordance between renal responses to captopril and to a renin inhibi- tor [19]. More recently, we found equivalent concordance to the renal vascular response to captopril and to the AT 1 receptor blocker candesartan in patients with type 1 diabetes mellitus [20]. Indeed, it was the findings in the latter study that led to the current analysis. The variations found in individual patients in this study are real, and may well reflect different stages in the process or different background pathophysiology. Although it was not our primary goal in this study to identify the factors responsible for a large renal hemody- namic response to blocking the RAS in some patients, and in minimal response in others, we did note that the nonresponders tended to be older, have a longer duration of diabetes mellitus, and to have remained com- plication free. One intriguing possibility is that these factors are mechanistically linked: Specifically, one might argue that the prolonged duration at risk without development of nephropathy is a product of the fact that they did not activate the RAS, and thereby were protected. An alternative possibility is that they escaped nephropa- thy for other reasons, and the RAS activation burned out during that long interval. While we find the former possibility to be the more attractive, no information in- patients probably reflects a reversal of angiotensin-mediated vasoconstriction, as suggested earlier [20]. What is new is the finding that the change in GFR induced by captopril and by candesartan was also highly concordant, suggesting that these changes are equally real. When captopril induced a sharp increase in GFR, so did candesartan. The equally strong concordance between change in RPF and change in GFR also induced by both captopril and candesartan is supportive of this hypothesis, and suggests a responsible mechanism: In the setting of this study, GFR was to a major degree RPF dependent. Renal plasma flow dependency of GFR, which occurs in some circumstances but not all, has been understood for thirty years [24 26]. Glomerular filtration occurs because hydrostatic pressure in the glomerular capillaries exceeds the offsetting oncotic pressure due to intravascular plasma protein. As filtration occurs, protein concentration in the glomerular capillaries rises, and filtration will cease when oncotic pressure equals glomerular capillary hydrostatic pressure. This equality between the two offsetting forces has been referred to as filtration pressure equilibrium. This state, which has been observed in the hydropenic rat [26] and hydropenic monkey [27], is a situation in which an increase in glomerular plasma flow rate, in the absence of significant changes in protein oncotic or capillary hydrostatic pressure, will result in a proportional rise in GFR. This rise occurs because an increase in plasma flow, in the absence of other changes in the determinants of GFR, results in a reduction in the rate of increased plasma protein concentration. As a consequence, the point at which filtration equilibrium is achieved is moved further toward the efferent end of the glomerular capillary network, effectively increasing the total capillary surface area exposed to a positive ultrafiltration pressure. The striking concordance between the change in RPF induced by both candesartan and captopril, and what is probably the consequent change in GFR, suggests strongly that the changes in GFR reflect RPF dependency. Whether or not filtration equilibrium exists appears to vary both with physiological state and the species. In the dog, filtration disequilibrium is the usual finding, and GFR varies little with RPF, for example, during plasma volume expansion [28]. In the hydropenic rat, as indicated earlier, filtration equilibrium is the routine. With volume expansion, on the other hand, at least some rats achieve filtration disequilibrium [17]. In humans, of course, only indirect evidence is available. More than sixty years ago, Homer Smith and coworkers suggested that filtration equilibrium was the usual situation in the healthy human kidney [29], a conclusion not confirmed by more recent analyses [30]. In accord with the latter, volume expansion and blocking the renin-angiotensin system (RAS) with ACE inhibitors or angiotensin receptor blockers will increase RPF in healthy humans but have little

6 Hollenberg et al: Glomerular dynamics 177 trinsic to this study would allow us to choose. Three at the tissue level. As the RAS clearly contributes to the major clinical subsets of patients, involving whether an development and progression of nephropathy in diabe- enhanced renal vascular response was present; whether tes, identification of the sources of the intrarenal RAS hypertension was present; and whether microalbuminuria was present was examined in this study, but as only activation has a substantial priority. five or six patients fell into these subsets, any conclusion ACKNOWLEDGMENTS must be tentative. We plan to revisit these issues when This work was supported in part by the National Institutes of Health our experience has increased two- to threefold. grants (NCRR GCRC M01RR02635, Hypertension SCOR 5P50HL Angiotensin-converting enzyme inhibition is a phar , Hypertension Training T32HL07609, and 1R01DK ). We are also grateful for the pharmaceutical support of AstraZeneca macologically more complex situation than angiotensin Pharmaceuticals, Molndal, Sweden. We thank Ms. Charlene Malarick, receptor blockade. ACE is also kininase-ii, and conseassistance in various aspects of this study. This research study was R.N., Ms. Caroline Coletti, B.S., M.S., and Ms. Diana Capone for their quently bradykinin levels can rise during ACE inhibition performed at Brigham and Women s Hospital on the General Clinical [33]. In a number of animal models, it has been suggested Research Center. that ACE inhibitor-induced bradykinin accumulation contributes to the renal hemodynamic response [7 11]. Reprint requests to Norman K. Hollenberg, M.D., Ph.D., Department of Medicine, Brigham and Women s Hospital, 75 Francis Street, Boston, There is evidence of striking species variation in the Massachusetts 02115, USA. contribution of bradykinin to renal hemodynamic responses, and in humans the findings are mostly negative djpagecapo@rics.bwh.harvard.edu [33, 34]. This study is in accord, as the renal hemody- REFERENCES namic response to candesartan was numerically larger 1. Hostetter TH, Troy JL, Brenner BM: Glomerular hemodynamics than the response to captopril, and there was no evidence in experimental insulin-dependent diabetes mellitus. Kidney Int of a qualitative or quantitative difference in the GFR 19: , Elving LD, Wetzels JMF, Denobel E, et al: Captopril reduces response to the two blockers. albminuria probably through lower of intraglomerular pressure: In our study, only a single dose of each of the pharma- Acute effects in patients with diabetic nephropathy. Postgrad Med cologic agents was employed. Our goal was to use the J (Suppl 64):34 39, Lewis EJ, Hunsicker LG, Bain RP, et al: The effect of angiotensinagents as probes to explore the underlying physiology converting enzyme inhibition on diabetic nephropathy. N Engl J and pathophysiology. The long-term therapeutic impli- Med 329: , 1999 cations would depend on the long-term response. The 4. Bakris GL, Weir MR: Angiotensin-converting enzyme inhibitor- associated elevations in serum creatinine: Is this a cause for confact that these agents are so very effective in patients cern? Arch Intern Med 160: , 2000 with type 1 diabetes and nephropathy [3] suggests that 5. ter Wee PM, Donker ABJM: Pharmacologic manipulation of glo- in some way the acute and the chronic renal response is merular function. Kidney Int 45: , Apperloo AJ, De Zeeuw D, De Jong PE: A short-term antihyperrelated. tensive treatment-induced fall in glomerular filtration rate predicts The rationale for the sequence, involving the adminis- long-term stability of renal function. Kidney Int 51: , 1997 tration of captopril on the first day and candesartan on 7. Clappison BH, Anderson WP, Johnston CI: Renal hemodynam- ics and renal kinins after angiotensin converting enzyme inhibition. the second, was described in an earlier report [20]. Our Kidney Int 20: , 1981 goal was to examine responses to the two agents with 8. Kon V, Fogo A, Ichikawa I: Bradykinin causes selective efferent the shortest possible interval. As candesartan has a protion. Kidney Int 44: , 1993 arteriolar dilation during angiotensin I converting enzyme inhibi- longed action, exceeding 48 hours in the human kidney, 9. Endlich K, Steinhausen M: Role of kinins and angiotensin II in and the action of captopril rarely lasts more than six to the vasodilating action of angiotensin converting enzyme inhibition eight hours, we chose not to randomize the sequence. in rat renal vessels. J Hypertens 15: , Heller J, Kramer HJ, Horacek V: Comparative effects of the In defense of this approach, in this study both RPF and angiotensin II receptor blocker EXP 3174 and of the angiotensin PRA had returned to baseline values 24 hours after cap- converting enzyme inhibitor captopril on renal glomerular hemodynamics in the dog. Kidney Blood Res 20: , 1997 topril administration. 11. Matsuda H, Hayashi K, Arakawa K, et al: Zonal heterogeneity Although the focus in this study was on intimate details in action of angiotensin converting enzyme inhibitor on renal miof the renal hemodynamic state, one overall finding is crocirculation role of intrarenal bradykinin. J Am Soc Nephrol of substantial interest. In this unselected group of 31 10: , Price DA, Porter LE, Gordon M, et al: The paradox of the lowpatients the intrarenal RAS was activated in over 80%, renin state in diabetic nephropathy. J Am Soc Nephrol 10:2382 as evidenced by a marked renal vasodilator response to 2391, 1999 both captopril and to candesartan. Although hyperglyceeffect of irbesartan on the development of diabetic nephropathy 13. Parving H-H, Lehnert H, Brochner-Mortensen J, et al: The mia probably contributed in at least some patients, it is in patients with type 2 diabetes. Irbesartan in Patients with Type clear that factors beyond glycemic state contribute to 2 Diabetes and Microalbuminuria Study Group. N Engl J Med the renal hemodynamic alterations. Blood sugar was 345: , 2001 maintained in the 100 mg/dl range with an insulin clamp. 14. Lewis EJ, Hunsicker LG, Clarke WR, et al: Renoprotective effect As PRA was increased in only a small percentage, it is clear also that the activation of the RAS occurs primarily of the angiotensin-receptor antagonist irbesartan in patients with nephropathy due to type 2 diabetes. Collaborative Study Group. N Engl J Med 345: , 2001

7 178 Hollenberg et al: Glomerular dynamics 15. Brenner BM, Cooper ME, De Zeeuw D, et al: Effects of losartan 25. Brenner BM, Troy JL, Daugharty TM, et al: Dynamics of glomerular on renal and cardiovascular outcomes in patients with type 2 diabetes ultrafiltration in the rat. II. Plasma-flow dependence of GFR. and nephropathy. RENAAL Study. N Engl J Med 345: , Am J Physiol 223: , Brenner BM, Troy JL, Daugharty TM: The dynamics of glomeru- 16. Carmines PK, Perry MD, Hazelrig JB, Navar LG: Effects of lar ultrafiltration in the rat. J Clin Invest 50: , 1971 preglomerular vascular resistance alterations on filtration fraction. 27. Maddox DA, Deen WM, Brenner BM: Dynamics of glomerular Kidney Int 31(Suppl 20):S229 S232, 1987 ultrafiltration. VI. Studies in the primate. Kidney Int 5: , Maddox DA, Brenner BM: Glomerular ultrafiltration (chapt 6), 28. Thomas CE, Bell PD, Navar LG: Glomerular filtration dynamics in The Kidney (vol 1, 4th ed), edited by Brenner BM, Rector in the dog during elevated plasma colloid osmotic pressure. Kidney FC, Philadelphia, W.B. Saunders Company, 1991, pp Int 15:502, Wesson LG: (editor): Physiology of the Human Kidney. New York, 29. Smith HW, Chasis H, Goldring W, Ranges HA: Glomerular Grune & Stratton, 1969, pp dynamics in the normal human kidney. J Clin Invest 19: , 19. Fisher NDL, Allan D, Kifor I, et al: Responses to converting 1940 enzyme and renin inhibition. Role of angiotensin II in humans. 30. Oken DE: An analysis of glomerular dynamics in rat, dog, and Hypertension 23:44 51, 1994 man. Kidney Int 22: , Lansang MC, Price DA, Laffel LMB, et al: Renal vascular responses 31. Hollenberg NK, Williams GH, Taub KJ, et al: Renal vascular to captopril and to candesartan in patients with type 1 response to interruption of the renin-angiotensin system in normal diabetes mellitus. Kidney Int 59: , 2001 man. Kidney Int 12: , Report on the Expert Committee on the Diagnosis and Classifica- 32. Miller JA: Impact of hyperglycemia on the renin angiotensin tion of Diabetes Mellitus. Diabetes Care 20: , 1997 system in early human type 1 diabetes mellitus. J Am Soc Nephrol 22. Emmanuel RL, Cain JP, Williams GH: Double antibody radioim- 10: , 1999 munoassay of renin activity and angiotensin II in human peripheral 33. Gainer JV, Morrow JD, Loveland A, et al: Effect of bradykinin- plasma. J Lab Clin Med 81: , 1973 receptor blockade on the response to angiotensin-converting-enzyme 23. Underwood RH, Williams GH: The simultaneous measurement inhibitor in normotensive and hypertensive subjects. N Engl of aldosterone, cortisol, and corticosterone in human peripheral J Med 339: , 1998 plasma by displacement analysis. J Lab Clin Med 79: , Lansang MC, Hollenberg: ACE inhibition and the kidney: spe- 24. Deen WM, Robertson CR, Brenner BM: A model of glomerular cies variation in the mechanisms responsible for the renal hemodynamic ultrafiltration in the rat. Am J Physiol 223: , 1972 response. JRAAS 1: , 2000