Diabetes

Diabetes, nephropathy, and the renin system
[Diabetes and renal effects]

Abstract

Blockade of the renin–angiotensin system has become crucial in the management of type 1 (insulin-dependent) and type 2 (non-insulin-dependent) diabetes mellitus, especially in patients who are at risk of nephropathy. In this review, we address the issue of why the renin system and its blockade are so important. As in many complex processes, diabetic nephropathy reflects an interaction between genetic factors and environmental factors. Recent research has uncovered a number of environmental factors; control of these factors should contribute to improved management. The renin system is important in patients with diabetes mellitus because so many relevant factors converge on the intrarenal renin system.

In studies that range from therapeutics to pharmacology to genetics, several lines of evidence have pointed to a role for the renin–angiotensin system (RAS) in the pathogenesis of nephropathy in patients with diabetes mellitus [1–6]. In 1990, Bjorck [3] reported on 18 studies that examined the effect of angiotensin-converting enzyme (ACE) inhibition on the renal blood supply in diabetes mellitus. He reported a consistent and striking renovasodilator response to ACE inhibition in type 1 diabetes mellitus. Only two of the 18 studies in Bjorck's report involved patients with type 2 diabetes mellitus, and sodium and potassium intake was controlled in none of these studies.

In the 15 years since Bjorck's review, both ACE inhibition [1] and angiotensin II (Ang II) type 1 receptor blocking agents (ARBs) [4–6] have become standard treatments in the prevention and management of nephropathy in the patient with diabetes mellitus. In this review, I will examine the state of the intrarenal RAS in both type 1 and type 2 diabetes mellitus, explore renal responses to ACE inhibition and ARBs, and review current understanding of the determinants of the state of the renin system and, thus, the risk of nephropathy. All these topics are immediately germane to therapeutic efficacy.

By 1990, it was appreciated that sodium intake and potassium intake have a sufficiently large influence on renal perfusion and angiotensin-dependent control that both should be controlled in renal hemodynamic studies. Unfortunately, none of the 18 studies reported until then [3] controlled either. De'Oliveira et al. [7] undertook a systematic assessment of dietary sodium and the renal hemodynamic response to ACE inhibition with enalapril [7]. Studies on the renal hemodynamic response to ACE inhibitors or ARBs have typically been best performed in individuals receiving a low-salt diet, to activate the renin system and increase the response [8]. All 19 participants in the study by De'Oliveira et al. [7] had their first renal hemodynamic study when in balance on a reduced (10 mmol) sodium intake. After that baseline study, nine remained on the low-salt diet, and 10 were switched to a diet that contained 200 mmol NaCl, but was otherwise unchanged. After 5–7 days when external sodium balance had been achieved, the renal hemodynamic and hormonal assessments were repeated. Responses to Ang II infused intravenously and to enalapril in doses up to 20 mg given daily for 3 days were also measured. The 19 individuals with diabetes were compared with 25 age-matched normal individuals.

In individuals with type 2 diabetes who were receiving a high-salt diet, enalapril increased renal plasma flow (RPF) from 477 ± 25 ml/min per 1.73 m² to 511 ± 29 ml/min per 1.73 m² (P = 0.03). The baseline response to Ang II was blunted in these patients with type 2 diabetes; enalapril also reversed this blunting, enhancing the renal vasoconstrictor response to Ang II from -68 ± 9 ml/1.73m² to -106 ± 18 ml/1.73m² (P = 0.03). Conversely, all measures made in individuals receiving a low-salt diet were normal in those with diabetes (Table 1). As the enhanced renal response to ACE inhibition was associated with enhancement of the renal vasoconstrictor response to Ang II, the findings suggested strongly that the renal response to ACE inhibition reflected reversal of Ang-II-mediated vasoconstriction. Surprisingly, this same pattern was evident even in the subset with low-renin hypertension. The findings suggested an autonomous RAS, which was suppressed abnormally to a lesser extent by a high-salt diet, despite greater volume expansion. Moreover, the state of the renin system was normal during the stimulation provided by a low-salt intake.

For several reasons, based on the findings of the study by De'Oliveira et al. [7], in diabetic individuals it has been attractive to perform the renal hemodynamic and RAS evaluations while they received a high-salt diet. First, that is clearly the most convenient approach, and spares a good deal of time. Secondly, as the renin system is suppressed by a high-salt diet in normal individuals and much less so in diabetic individuals, the high-salt procedure increases the difference between the two groups. Thirdly, and perhaps scientifically most important, the continuing glycosuria and consequent osmotic diuresis in the individuals with diabetes leads to continued loss of sodium in the urine, and thus exaggerates the effects of a low-salt diet in patients in whom glycosuria could not be controlled. Glycosuria was minimal or absent in the patients included in the study by De'Oliveira et al. [7].

Angiotensin-converting enzyme inhibition and the kidney

Given the large number of studies that have revealed an enhanced renal vasodilator response to inhibition of the renin system in both type 1 and type 2 diabetes mellitus [3,7], the fact of an enhanced renal vascular response appeared to be well established. In contrast, the mechanism was not. One possibility involved a reduction in formation of Ang II, but there was an alternative possibility that involved activation of vasodilator pathways via the accumulation of bradykinin, the formation of prostaglandin, and the activation of nitric oxide synthesis [9]. Many of the studies suggesting activation of a vasodilator pathway in the kidney were performed in animal models. Lansang and Hollenberg [10] performed a blinded assessment of these studies and found that the evidence was very powerful in the dog and the rat, but much less so in the rabbit or humans. An important question was whether humans really were more like the rabbit or more like the dog and rat. Species variation has been very important in the contribution of the renin system to intrarenal control mechanisms. The blunting of renovascular responses to Ang II at baseline in the diabetic patients in the study by De'Oliveira et al. [7] and the enhancement of that response after ACE inhibition suggested that the primary mechanism for renal vasodilatation in response to enalapril involved a reduction in Ang II. An increase in bradykinin or nitric oxide would have further blunted the response to Ang II. Although compelling, this evidence was not definitive. Thus we sought an alternative approach, which we found in the assessment of responses to candesartan and to captopril.

In a study designed to assess the mechanism responsible for the accentuated renal hemodynamic response, 12 men and women with type 1 diabetes were studied [11]. Each was studied twice on a single admission. On one study day, they received captopril 25 mg. On a second study day, they received candesartan 16 mg. These doses were chosen as they represent the top of the relationship between dose and renovascular response.

The responses to captopril were striking (Fig. 1). In five of the 12 individuals studied, the renal vasodilator response was minimal, compatible with the normal range of responses. In the other seven, the response was enhanced substantially compared with normal individuals studied while receiving a high-salt diet. Immediately relevant to the issues raised in this review was the extraordinarily concordant correlation between the two responses (Fig. 1; r = 0.86; P < 0.001). Every individual with a very large response to captopril also had a very large response to candesartan; of the five with a minimal response to captopril, four showed a minimal response to candesartan. Rarely in medicine do we find correlation coefficients of 0.86, unless we are calibrating a system. More than 75% of the response to captopril could be accounted for on the basis of the response to candesartan. This concordance indicates that reduced formation of Ang II is the dominant mechanism of action of the ACE inhibitor. In turn, we believe that this action reflects an activated baseline state of the intrarenal renin system – despite a normal or low plasma renin activity.

This interpretation has been challenged on a number of counts. The most common involves the question, ‘How do you know that it isn't responsiveness to Ang II that is involved, rather than the amount of Ang II in the kidneys?’. The answer is straightforward. We have found a blunted response to Ang II in patients with diabetes [7], and that response is enhanced by ACE inhibition. Thus an enhanced response to Ang II cannot account for the enhanced response to ACE inhibition and ARBs. A second question is, ‘How do you know that it isn't a response to the blood pressure changes induced by the ACE or the ARB?’. In fact, a significant depressor response to ACE inhibition or ARBs in a single dose administered to a patient receiving a high-salt diet is rare. Thirdly, ‘How do you know that the response to both the ACE inhibitor and the ARB does not reflect some third system at the microvascular level?’. Evidence based on a drug is always somewhat suspect, as few drugs enjoy a single action. However, the pharmacologies of ACE inhibition and ARBs overlap, it appears, only at the intersection involving the generation or effect of Ang II. Even if there is a third mechanism that participates, we believe it extremely likely that the RAS is involved. Despite the indirectness of the methodology, this approach provides compelling data, and there seem to be few alternative approaches applicable in humans.

Control of glomerular filtration rate in diabetes mellitus

On average, the glomerular filtration rate (GFR) did not change in many of these earlier studies [3,7,11]. In contrast, we did notice a pattern of GFR responses, which was pursued in a larger study involving 31 patients with type 1 diabetes mellitus, designed to address several questions [12]. First, what is the frequency of an enhanced intrarenal RAS in patients with diabetes, based on the criterion of an enhanced renal vasodilator response to captopril and to candesartan? Secondly, what was the pattern of GFR response in relation to the renal vasodilator response? Thirdly, what could we learn about the nature of intrarenal hemodynamics from an analysis of these relationships?

There was a wide range of changes in RPF and GFR in response to these two agents, each administered at the top of its dose–response range. As previously, the RPF response to the two agents was strongly concordant (r = 0.65; P < 0.001; Fig. 2a). On average, GFR, once again, did not change. However, there was a strong correlation between the GFR response to captopril and that to candesartan (Fig. 2b). Thus those patients in whom GFR increased in response to captopril showed a similar increase in response to candesartan. Those patients in whom GFR did not increase, or in whom it decreased, with captopril also showed a decrease with candesartan (r = 0.81; P < 0.001).

The mechanism responsible for the change in GFR became evident when the renal vasodilator response to each agent reflected by RPF was examined in relation to the GFR response (Fig. 3). Large increases in RPF with candesartan were associated with a substantial increase in GFR (r = 0.83; P < 0.001). This is clear evidence of the RPF-dependency of the GFR in this situation. The relationship for the responses to captopril was similar (Fig. 3a).

Why have these relationships been missed in earlier studies, and what do they mean? In general, the renal hemodynamic response to ACE inhibition or ARBs in renal disease or in patients at risk of renal injury has been presented as group means. When examined in this way, we found in this current study that, as others have reported in the past, RPF increased, and GFR on average did not change. As a consequence, filtration fraction decreased. The variations found in the renal hemodynamic pattern of individual patients are real and relevant to their disease. They may well reflect different stages in the process or different background pathophysiology, including a different genetic component.

About 80% of the patients showed an enhanced response. A tendency was noted that the non-responders – that is those in whom the response was in the normal range – tended to be older, to have a longer duration of diabetes mellitus, and, despite this, to have remained free of complications. One intriguing possibility is that these factors are mechanistically linked: specifically, one might consider that the prolonged duration of risk without development of nephropathy is a product of the fact that these individuals do not activate the RAS, and thereby are somewhat protected. An alternative possibility is that they escape nephropathy for other reasons and that activation of the RAS had ‘burnt out’ during the long period. Perhaps genetics will help us to resolve this issue [2,13–15].

The mechanisms responsible for GFR have been understood for decades [12]. Glomerular filtration occurs because hydrostatic pressure in the glomerular capillaries exceeds the offsetting oncotic pressure caused by intravascular plasma protein. As filtration occurs, protein concentration in the glomerular capillaries increases: filtration ceases when oncotic pressure equals glomerular capillary hydrostatic pressure. This equality between the two offsetting forces has been referred to as ‘filtration pressure equilibrium’. This is a situation in which an increase in glomerular plasma flow rate, in the absence of significant change in protein oncotic pressure, will result in a proportional increase in GFR, because the increase in plasma flow results in a reduction in the rate of increased plasma protein concentration in the glomerular capillaries. 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 available capillary surface area exposed to a positive ultrafiltration pressure. The striking concordance between change in RPF induced by both candesartan and captopril and what is probably the consequent change in GFR suggests strongly that changes in GFR in these patients reflect RPF dependency.

With the development of diabetes mellitus, there is a substantial increase in both renal and glomerular size, along with concomitant changes in glomerular capillaries. Our findings raise the interesting possibility that one physiological 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. For these reasons, filtration fraction does not provide an index of glomerular capillary pressure in the patient with diabetes mellitus, an observation that should bother many nephrologists – who should have known better!

Given the evidence that activation of the intrarenal RAS contributes to the pathogenesis of diabetic nephropathy, a model that allows us to assess the state of the intrarenal renin system could provide insight into the risk of nephropathy. There are a number of identified risk factors, which include specific genes, obesity, ethnicity, sex, and – most recently – the identification of oral contraceptive use in young women with type 1 diabetes mellitus.

Diabetic nephropathy has provided a very useful model for exploring the interaction between genetic factors and factors in the environment [13]. Nephropathy occurs in 30–40% of patients, and diabetes is the most common cause of end-stage renal disease. Although hyperglycemia and hypertension clearly contribute to nephropathy, they do not fully explain the development of nephropathy in patients with diabetes. Family studies have revealed familial clustering in both type 1 and type 2 diabetes, with about a threefold increase in risk in siblings if the proband had diabetic nephropathy. Krolewski reviewed evidence to suggest a major gene effect, based on the assessment of sibling pairs [14]. If the first sibling in a sibling pair has diabetes mellitus and nephropathy, there is a more than 70% likelihood that the second sibling with diabetes mellitus will also develop nephropathy. Conversely, if the first sibling in the pair is free of nephropathy, there is only about a 25% likelihood that the second sibling will have nephropathy. This threefold increase in risk identified on the basis of the first sibling is not compatible with polygenic inheritance. There must be a contribution from one or more major genes.

Given the contribution of the renin system to diabetic nephropathy, it was logical that the search for candidate genes that might contribute led to studies centering on polymorphisms involving the genes governing the RAS. Despite an enormous number of studies, the results remain equivocal. This is the situation in which an intermediate phenotype has proven to be very helpful. An intermediate phenotype is a physiological process that conceptually links the influence of the gene on the one hand, and disease on the other. We are in the process of testing the hypothesis that the renal response to captopril and ARBs will provide a useful intermediate phenotype in the assessment of the contribution of specific genes.

Obesity is a common feature in diabetes mellitus, especially in type 2 diabetes. We have long recognized that adipose tissue synthesizes angiotensinogen, the substrate for the renin reaction [16]. The plasma concentration of angiotensinogen is increased in obese patients [17]. Some years ago, we recognized that obesity was associated with blunting of the renovascular response to Ang II administered intravenously, in association with a polymorphism of the angiotensinogen (AGT) gene [15]. This observation prompted two lines of investigation. In one, we examined the renal vascular response to blocking the RAS with an ARB in patients with type 2 diabetes mellitus [18]. We found that body mass index (BMI), used as a measure of obesity, was highly correlated with the renal response to the ARB (r = 0.7; P = 0.01). Thus about 50% of the enhanced response to ARB reflected obesity. Our findings suggest an important role for obesity in activating the intrarenal renin system, perhaps via production of angiotensinogen. BMI must be considered a risk factor for nephropathy, and we should consider recommending weight loss as part of management.

In our second line of investigation, we examined the renal hemodynamic response to captopril and to candesartan in healthy normal individuals over a wide range of BMI. Again, there was a strong positive correlation between BMI and the renal vasodilator response [19]. Of some interest was the fact that a quadratic relationship provided the best fit for the data (r = 0.55; P < 0.001). We suspect that this reflects a threshold effect. There was essentially no response in individuals with a BMI less than 26 kg/m². In the BMI range 26–30 kg/m², the response became significant; in those with a BMI exceeding 30 kg/m², the enhancement of a response accelerated rapidly.

Ethnicity is another risk factor for diabetic nephropathy. Black persons with type 2 diabetes have a four- to 10-fold increase in the risk of nephropathy compared with white individuals [20]. We have found that, in apparently healthy black individuals, free of hypertension or diabetes, the contribution of the RAS to renal hemodynamic status differs from that in their white counterparts [21,22]. In association with a high-salt diet, RPF is lower in black than in white individuals, and the renal vascular response to ACE inhibition with captopril is enhanced substantially. Conversely, with a low-salt diet, renal perfusion is identical in black and white individuals, and the response to ACE inhibition and ARBs is also identical in the two groups. Thus, as in those with diabetes, the ethnic difference reflects, not an overly active RAS in the kidney, but rather a more limited suppression of intrarenal renin on a high-salt intake. As another expression of the same physiological difference, RPF does not change on a shift from a low-salt to a high-salt diet in black individuals, whereas there is at least a 10% increase in RPF in white groups [22]. With this limited perfusion, black individuals show more sodium retention, gain more weight, and show a larger increase in blood pressure. If the intrarenal RAS plays a part in the pathogenesis of hypertension, specifically salt-sensitive hypertension and in the pathogenesis of diabetic nephropathy, then the healthy black person is at risk for reasons that are not yet clear. One possibility that demands examination involves the fact that black individuals, on average, have a greater BMI than whites.

Sex is another risk factor for diabetic nephropathy – and, indeed, all forms of nephropathy. Women are at substantially lower risk than men, for reasons that have remained obscure [23]. We noted many years ago that the first generation of oral contraceptive agents activated the renin system and induced renal vasoconstriction in apparently healthy young women [24]. Because hypertension is much less of a problem with third-generation oral contraceptive agents, they have not been considered a risk factor. We have recently demonstrated in healthy young women that third-generation oral contraceptive agents, like the first-generation agents, enhance the response of the renal blood supply to captopril and to candesartan [25]. In the same investigation, we took advantage of an inception cohort study performed by Hans-Henrik Parving and his coworkers in Denmark [25] to examine the influence of oral contraceptive agents on risk of nephropathy in young women with type 1 diabetes. We found, to our astonishment, that there was an extraordinary increase in risk of nephropathy in young women receiving an oral contraceptive agent, the increase in frequency of nephropathy being 900%! There are few risk factors in medicine to match a ninefold increase in risk.

Taking all the evidence together, it is difficult to ignore a massive implication of the intrarenal RAS as a risk factor, probably a pathogenetic factor, in the genesis of diabetic nephropathy. The fact that blocking the system with an ACE or an ARB reduces the risk of nephropathy or slows its progression is now beyond denial.

Optimization of renin–angiotensin system blockade

Our new challenge lies, not in establishing the efficacy of the approach, but rather in learning how to optimize renin system blockade [26]. Most attention has been given to ACE–ARB combinations. Many, but by no means all, of the ACE–ARB combination studies have shown an improvement in nephropathy compared with either agent alone [27]. Because the issue of dosage of the two agents has not been dealt with in an ideal manner, this remains a controversial area. As ARBs are so very well tolerated, an alternative approach based on very high doses of ARBs – well above the dose used for blood pressure control – has been receiving increasing attention. Currently, studies are in progress in patients with type 2 diabetes and proteinuria, to ascertain whether increased doses of three ARBs – up to 128 mg of candesartan daily, up to 900 mg of irbesartan daily, and up to 640 mg of valsartan daily – will improve proteinuria to a greater extent than can be achieved with standard doses. When these studies are completed, we will know how to design studies addressing the ACE–ARB combination more effectively. A third relevant issue involves how to use aldosterone antagonists along with ACE inhibitors or ARBs to improve natural history. Finally, renin inhibitors work at the rate-limiting step of the cascade, and their contribution to control of diabetic nephropathy also requires attention.

The past two decades have proved to be truly interesting. The next decade or two promise to be equally so.

1 Lewis EJ, Hunsicker LG, Bain RP, Rohde RD. The effect of angiotensin-converting enzyme inhibition on diabetic nephropathy. N Engl J Med 1993; 329:1456–1462. Bibliographic Links [Context Link]

2 Marre M, Bernadet P, Gallois Y, Savagner F, Guyene TT, Hallar M, et al. Relationship between Ang I converting enzyme gene polymorphism, plasma levels, and diabetic complications. Diabetes 1994; 43:384–388. [Context Link]

3 Bjorck S. The renin–angiotensin system in diabetes mellitus. A physiological and therapeutic study. Scand J Urol Nephrol Suppl 1990; 126:1–50. Bibliographic Links [Context Link]

4 Lewis EJ, Hunsicker LG, Clarke WR, Berl T, Pohl MA, Lewis JB, et al. Renoprotective effect of the angiotensin receptor antagonist irbesartan in patients with nephropathy due to type 2 diabetes. N Engl J Med 2001; 345:851–860. Bibliographic Links [Context Link]

5 Parving H-H, Lehnert H, Bröchner-Mortensen J, Gomis R, Andersen S, Amer P. The effect of irbesartan on the development of diabetic nephropathy in patients with type 2 diabetes. N Engl J Med 2001; 345:870–878. Bibliographic Links [Context Link]

6 Brenner BM, Cooper ME, De Zeeuw D, Keane WF, Mitch WE, Parving HH, et al. Effects of losartan on renal and cardiovascular outcomes in patients with type 2 diabetes. N Engl J Med 2001; 345:861–869. Bibliographic Links [Context Link]

7 De'Oliveira JM, Price DA, Fisher NDL, Allan DR, McKnight JA, Williams GH, Hollenberg NK. Autonomy of the renin system in type II diabetes mellitus. Dietary sodium and renal hemodynamic responses to ACE inhibition. Kidney Int 1997; 52:771–777. Bibliographic Links [Context Link]

8 Shoback DM, Williams GH, Hollenberg NK, Davies RO, Moore TJ, Dluhy RG. Endogenous Ang II as a determinant of sodium modulated changes in tissue responsiveness to Ang II in normal man. J Clin Endocrinol Metab 1983; 57:764–770. Bibliographic Links [Context Link]

9 Mori M, Akatsuka N, Fukazawa M, Hara K, Ishikawa M, Aikawa J, et al. Endothelin-dependent relaxation by angiotensin-converting enzyme inhibitors in canine femoral arteries. Am J Physiol 1994; 266:H583–H589. [Context Link]

10 Lansang MC, Hollenberg NK. ACE inhibition and the kidney. Species variation in the mechanisms responsible for the renal haemodynamic response. J Renin Angiotensin Aldosterone Syst 2000; 1:119–124. Bibliographic Links [Context Link]

11 Lansang MC, Price DA, Laffel LMB, Osei SY, Fisher NDL, Erani D, Hollenberg NK. Renal vascular responses to captopril and to candesartan in patients with type 1 diabetes mellitus. Kidney Int 2001; 59:1432–1438. Bibliographic Links [Context Link]

12 Hollenberg NK, Price DA, Fisher NDL, Lansang MC, Perkins B, Gordon MS, et al. Glomerular hemodynamics and the renin–angiotensin system in patients with type 1 diabetes mellitus. Kidney Int 2003; 63:172–178. Bibliographic Links [Context Link]

13 Lansang MC, Hollenberg NK. Genes, environment, intermediate phenotypes, and the pathogenesis of diabetic nephropathy. Curr Pharmacogenom 2004; 2:61–74. [Context Link]

14 Krolewski AS. Genetics of diabetic nephropathy: evidence for major and minor gene effects. Kidney Int 1999; 55:1582–1596. Bibliographic Links [Context Link]

15 Frederich RC Jr, Kahn BB, Peach MJ, Flier JS. Tissue-specific nutritional regulation of angiotensinogen in adipose tissue. Hypertension 1992; 19:339–344. Bibliographic Links [Context Link]

16 Umemura S, Nyui N, Tamura K, Hibi K, Yamaguchi S, Nakamaru M, et al. Plasma AGT concentrations in obese patients. Am J Hypertens 1997; 10:629–633. Bibliographic Links [Context Link]

17 Hopkins PN, Lifton RP, Hollenberg NK, Jeunemaitre X, Hallouin MC, Skuppin J, et al. Blunted renal vascular response to angiotensin II is associated with a common variant of the angiotensinogen gene and obesity. J Hypertens 1996; 14:199–207. Bibliographic Links [Context Link]

18 Price DA, Lansang MC, Osei SY, Fisher NDL, Laffel LMB, Hollenberg NK. Type 2 diabetes, obesity, and the renal response to blocking the renin system with irbesartan. Diabet Med 2002; 19:858–861. Bibliographic Links [Context Link]

19 Ahmed SB, Fisher NDL, Stevanovic R, Hollenberg NK. Body mass index and angiotensin-dependent control of the renal circulation in healthy humans. Hypertension 2005; 46:1316–1320. Bibliographic Links [Context Link]

20 United States Renal Disease Survey. Annual data report. Atlas of end-stage renal disease in the United States. Bethesda, MD: National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases; 2003. Report No. ?. [Context Link]

21 Price DA, Fisher NDL, Osei SY, Lansang MC, Hollenberg NK. Renal perfusion and function in healthy African Americans. Kidney Int 2001; 59:1037–1043. Bibliographic Links [Context Link]

22 Price DA, Fisher NDL, Lansang MC, Stevanovic R, Williams GH, Hollenberg NK. Renal perfusion in blacks. Alterations caused by insuppressibility of intrarenal renin with salt. Hypertension 2002; 40:186–189. Bibliographic Links [Context Link]

23 Kang AK, Miller JA. Impact of gender on renal disease. The role of the renin angiotensin system. Clin Invest Med 2003; 26:38–44. Bibliographic Links [Context Link]

24 Hollenberg NK, Williams GH, Burger BM, Chenitz WR, Hooshmand I, Adams DF. Renal blood flow and its response to angiotensin II. An interaction between oral contraceptive agents, sodium intake and the renin–angiotensin system in healthy young women. Circ Res 1976; 38:35–40. Bibliographic Links [Context Link]

25 Ahmed SB, Hovind P, Parving H-H, Rossing P, Price DA, Laffel LM, et al. Oral contraceptives, angiotensin-dependent renal vasoconstriction, and the risk of diabetic nephropathy. Diabetes Care 2005; 28:1988–1994. Bibliographic Links [Context Link]

26 Hollenberg NK. Treatment of the patient with diabetes mellitus and risk of nephropathy. What do we know, and what do we need to learn? Arch Intern Med 2004; 164:125–130. Bibliographic Links [Context Link]

27 Andersen NH, Mogensen E. Dual blockade of the renin angiotensin system in diabetic and nondiabetic kidney disease. Curr Hypertens Rep 2004; 6:369–376. Bibliographic Links [Context Link]