 |
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
|
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
Nephrology beyond JASN |
Renal failure in type 2 diabetes is truly a medical catastrophe of worldwide dimension (1): In all countries of the western world, diabetes, mostly type 2, has become the most frequent comorbid condition in patients admitted for renal replacement therapy. It had long been recognized that the classical features of diabetic nephropathy, so well documented in type 1 diabetes (2,3), are not present in patients with type 2 diabetes and impaired renal function in an equally uniform fashion. In the third NHANES survey, Kramer et al. (4) noted that retinopathy and albuminuria (spot urine albumin/creatinine ratio) were absent in 30% of elderly type 2 diabetic patients with eGFR <60 ml/min per 1.73 m2 (2) (Modification of Diet in Renal Disease [MDRD] formula). In several independent local surveys we found that 15 to 20% of type 2 diabetic patients reaching end-stage renal failure lacked major proteinuria and had shrunken kidneys, raising the unproven assumption of ischemic nephropathy (5). It is well known that in earlier stages of renal disease the renal histology is much less uniform in type 2 than in type 1 diabetes, with a high frequency of atypical patterns including tubulo-interstitial lesions, advanced glomerular hyalinosis, or global sclerosis (6,7).
In the past, the group from Heidelberg/Victoria reported a considerable prevalence of reduced GFR in the absence of albuminuria/proteinuria, but this study relied on the notoriously unreliable creatinine clearance values, and a confounding effect of treatment could also not be definitely excluded (8). The authors now present the results of a cross-sections survey using radiochelate clearance measurements as the gold standard and complement this with an analysis of longitudinal measurements in a small subgroup. They solidly confirm the surprisingly high prevalence of nonalbuminuric chronic kidney disease (CKD) with GFR <60 ml/min per 1.73m (2) in type 2 diabetes, although the prevalence in a matched sample of the background population (presumably much lower) remains unknown. In the 36% of the total sample of type 2 diabetic patients who had a GFR <60, the frequency of normo-, micro-, and macroalbuminuria was 39%, 35%, and 26%, respectively. Normoalbuminuria was found most frequently in elderly females. Even when patients with blockade of the renin-angiotensin system (which might have caused reversal to normoalbuminura) were excluded, the prevalence of normoalbuminuria still remained 23%.
In a very small subset of patients with impaired renal function, longitudinal GFR measurements were available. The rate of loss of GFR was certainly higher than expected in this age group and not significantly different between patients with normoalbuminuria (–4.6 ± 1.0 ml/min per 1.73 m2 [2]), microalbuminuria (–2.8 ± 1.0 ml/min per 1.73 m2) and macroalbuminuria (–3.0 ± 0.7 ml/min per 1.73 m2).
This study certainly does not invalidate the use of urine analysis for albumin/protein, nor does it disprove a pathogenetic role of proteinuria. Nevertheless it illustrates that matters are more complex than we thought. The suspicion is raised that additional or complementary pathomechanisms operate in a by no means small segment of elderly type 2 diabetic patients. Unfortunately we were not given information on potential vascular disease, on kidney size, and, above all, on the gold standard of renal biopsy. The latter is particularly relevant because, in normoalbuminuric type 1 diabetic patients with reduced renal function, advanced glomerular lesions were described by Caramori (9). Although in these type 2 diabetic patients this explanation is by no means excluded in the absence of renal biopsies, one should also consider the possibility that diabetes causes premature senescence of the kidney, excessive sensitization of renal vasculature to BP, or deterioration of renal function via interstitial fibrosis to the exclusion of advanced glomerular pathology. Alternatively the kidney may be the victim of extrarenal factors such as a history of hypertension, malnutrition, or cholesterol microembolism, to mention only a few. It will be particularly relevant to address the above potential nonclassical pathogenetic pathways if we are ever to arrive at specific targeted interventions for this segment of the diabetes population.
At any rate, we can draw one practical conclusion which is obvious to the nephrologists but may be alarming news to many non-nephrologists: in the type 2 diabetic patient it is not good enough to just examine the urine—it is also necessary to measure or estimate renal function.
|
Address correspondence to: Prof. Eberhard Ritz, Department Internal Medicine, Division of Nephrology, Bergheimer Strasse 56a, D-69115 Heidelberg, Germany. Phone: 49-0-6221-601705 or 49-0-6221-189976; Fax: 49-0-6221-603302; E-mail: Prof.E.Ritz{at}t-online.de
References
In the past, nephrologists were aware of the heritable disorders of renal phosphate wasting, i.e., X-linked hypophosphatemia (XLH) and the rare autosomal dominant hypophosphatemic rickets (ADHR), as well as tumor-induced osteomalacia (1), long considered exotic and rare conditions of limited practical interest. All this has dramatically changed with the unraveling of the underlying genetics. A mutation of a protease (PHEX) was identified in XLH and in ADHR, a mutation of fibroblast growth factor 23 (FGF23) located in a sequence motif crucial for cleavage by furin proteases. The exact role of PHEX is still controversial (2) and additional molecules of unknown pathogenetic significance have been identified, i.e., the phosphaturic frizzled related protein (FRP4) and matrix extracellular phosphoglycoprotein of unknown function (3). The crucial breakthrough that got this field moving has been the cloning and characterization of FGF23 as a causative factor of tumor-induced osteomalacia by Shimada et al. (4). Native as well as mutated FGF23 suppresses Na+-dependent phosphate cotransport in proximal tubules (5) and—somewhat unanticipated—also reduces the 1-
hydroxylase activity as well as circulating 1,25(OH)2D3 concentrations. Recombinant FGF23 carrying the ADHR mutation causes phosphaturia and hypophosphatemia (6), and conversely targeted ablation of the FGF23 gene causes hyperphosphatemia (7), documenting the key importance of this molecule in the maintenance of normal phosphate concentrations.
For nephrologists the scene has become exciting with the observation in animals and patients with impaired renal function that the concentrations of the phosphaturic "hormone" FGF23 are elevated (8–10), although some methodological points remain to be clarified. The unresolved issue remained: Was oversecretion of FGF23 just an appropriate response to phosphate retention, or was FGF23 a primary mover in renal failure provoking the known endocrine abnormalities of hyperparathyroidism and deficiency of active vitamin D? The latter possibility is rendered increasingly more plausible by recent studies. Transgenic mice expressing FGF23 under the control of an 1(I)collagen promoter in bone had low serum-P concentrations, high urinary P excretion, and high parathyroid hormone (PTH) concentrations, while there were no differences in serum creatinine and 1,25(OH)2D3 (11). Even more suggestive are the results in a model where FGF23 was overexpressed in the liver under the influence of apolipoprotein E3 as a promoter. FGF23 carried a mutation (R176Q) which protected against enzymatic degradation. Such massive overproduction of less biodegradable FGF23 caused not only hypophosphatemia and phosphaturia, but also low 1,25(OH)2D3 and massively elevated PTH concentrations associated with severe bone lesions (rickets and osteomalacia). The observation that high FGF23 causes hyperparathyroidism in the absence of renal malfunction, either directly or indirectly via lower 1,25(OH)2D3 concentrations, is obviously tantalizing in view of the question whether high FGF23 concentrations in renal failure are just an innocent reaction to hyperphosphatemia or driving the endocrine abnormalities of PTH oversecretion and 1,25(OH)2D3 underproduction. In this context it is of great interest that long ago high PTH had been noted in patients with ADHR despite normal renal function and before any therapeutic administration of phosphate (12,13). The same was seen in XLH (14).
This is a rapidly moving frontier and the nephrologist is well advised to stay tuned. It is too early in the day to draw a definite pathogenetic scheme, but it is likely that in the genesis of hyperparathyroidism of renal failure, the classical trio of low Ca, high P, and low active D will now become a quartet including FGF23 as an active player.
References
In chronic renal disease, iron metabolism is often deranged and functional iron deficiency with low transferrin saturation but high serum ferritin concentration is frequent. This poses diagnostic dilemmas as illustrated by the observation that liver iron content measured by superconducting quantum interference device (SQUID) is very poorly correlated to the serum ferritin concentration (1). Abnormal regulation of iron metabolism is also thought to contribute to anemia and reduced responsiveness to erythropoietin.
In the past, a vexing issue had been which pathomechanisms account for the disturbed iron metabolism in patients with anemia of chronic disease. Anemia of chronic disease is generally normocytic and normochromic, but may become hypochromic or microcytic; it tends to be associated with low serum iron and and high serum ferritin, low transferrin saturation, shortened red cell survival, and hyporesponsiveness to erythropoietin, features that are also common in chronic renal disease. Anemia of chronic disease is characterized by increased storage of tissue iron, but at the same time by diminished iron release from the reticuloendothelial system (RES) and by diminished iron absorption from the intestine, resulting in diminished availability of iron for erythropoiesis.
The recent work of Nemeth et al. (2) goes a long way toward explaining the pathomechanisms underlying the disturbed iron metabolism in the anemia of chronic disease by documenting that IL-6 stimulates synthesis and secretion of hepcidin, a negative regulator of iron metabolism. Why is this observation important and potentially relevant for the understanding of the disturbed handling of iron in renal disease? Hepcidin inhibits the release of iron from macrophages and from cells of the RES (3), its absorption from the intestine and its transplacental passage (4,5). This leads to decreased serum iron concentrations and reduced availability of iron for erythropoiesis. The finding of Nemeth et al. that inflammation causes increased hepcidin synthesis, secretion and excretion in the urine is presumably also relevant for chronic renal failure, a known state of microinflammation and deranged iron metabolism. This is of obvious interest to the nephrologist (6), not in the least because the known constellation of intestinal iron malabsorption (7) and tissue sequestration of iron (1), so commonly found in renal patients, closely resemble the abnormalities of iron handling seen in inflammatory states.
What Is Hepcidin?
Hepcidin’s discovery is a beautiful example of how unpredictable the ultimate results of scientific investigation can be. In 2000 Krause et al. screened human blood ultrafiltrate for antimicrobial peptides. They isolated and characterized a 25–amino-acid peptide with eight disulfide-bonded cystein residues, a member of the defensin family, which he called LEAP (liver-expressed antimicrobial peptide) (8). The cDNA sequence predicted an 84–amino-acid prepropeptide with two consensus cleavage sites. The same peptide was also isolated from human urine and called "hepcidin," i.e., a bactericidal substance of hepatic origin (9).
It turns out that hepcidin is not a bactericidal substance alone (although presumably not bactericidal in the blood because the bactericidal properties are abrogated by 140 mmol/L NaCl). Hepcidin also restricts the availability of iron for erythropoiesis and other iron-dependent processes by inhibiting intestinal iron absorption and by promoting its sequestration in the RES—a host defense mechanism which makes a lot of sense biologically (10), because bacteria need iron among other things for the production of superoxide dismutase required for the defense against oxygen radicals produced by the host. This may explain why hepcidin has been strongly preserved in evolution down to fish and insects (11–13).
Hepcidin is the long-sought central regulator of iron metabolism (14) that plays a crucial role in iron homeostasis. This is suggested by the observation of severe iron overload in mice lacking hepcidin (15) and conversely by fatally severe iron deficiency of transgenic mice overexpressing hepcidin (5).
Similarly in humans, loss of function mutations of hepcidin cause one form of severe juvenile hemochromatosis (16). Conversely, a study in patients with autonomous hepcidin overxpression in hepatic adenomas documented that increased hepcidin production causes severe iron-resistant hypochromic anemia, which is reversible after resection of the hepcidin-producing adenoma (17).
How Is This Important Substance Regulated?
The work of Nicolas (18) addressed the effects of blood loss (phlebotomy), hypoxia (hypobaric chamber), and inflammation (injection of turpentine) on hepcidin expression. Anemia and hypoxia caused decreased hepcidin expression in the liver, a biologically useful response, because rapid mobilization of iron from the RES (and later iron absorption from the intestine) provide iron for erythropoiesis, thus correcting the stimuli of anemia and hypoxia, respectively. The response to the inflammatory stimulus was remarkable. A single injection of turpentine caused a decrease of serum iron in wild-type mice (but not in hepcidin knock-out mice) as well as a dramatic increase of hepcidin mRNA in the liver, analogous to the effect of LPS in mice (19). In agreement with this experimental finding, an increased excretion of hepcidin in the urine was also seen in patients with anemia of chronic disease secondary to inflammatory disorders and infections (20)
The study of Nemeth (2) now carries this issue one step further by showing, in human liver cell cultures, in mice, and in human volunteers, that IL-6 is the necessary and sufficient cytokine for the induction of hepcidin production and decreased serum iron concentration provoked by inflammation.
In the model of inflammation by turpentine injection, a decrease in serum iron concentration and increased hepcidin expression in the liver were observed in wild-type mice, but not in IL-6 knock-out mice, underlining the crucial role of IL-6 for the hepcidin response to inflammation.
The salient observation was made in human volunteers. IL-6 infusion caused a rapid, major increase of hepcidin excretion in the urine. This was paralleled by a decrease in serum iron and in transferrin saturation.
In contrast, dietary iron caused a rapid increase of hepcidin excretion in study subjects, but in an ancillary experimental study such increase was also seen in IL-6 knock-out mice, indicating that the hepcidin response to the stimulus of iron load is IL-6–independent.
This field is rapidly moving and many questions are currently unresolved. Injection of erythropoietin into mice caused a dramatic decrease of hepcidin expression in the liver (21), but it remains unclear whether this is due to a direct effect of erythropoietin or an indirect effect of consumption of iron for erythropoiesis. It also remains to be seen whether erythropoietin is directly involved in the downregulation of hepcidin in response to hypoxia and anemia.
An obvious question is whether hepcidin concentrations are altered in renal patients. In two studies (22,23) increased prohepcidin concentrations were found in renal patients, but the methodology is still not completely satisfactory and, above all, prohepcidin is presumably not the active agent. Hepcidin currently can not be measured in blood and measurement of urinary hepcidin remains the methodological gold standard (2).
Are there perspectives for diagnosis and therapy? It is a suggestive possibility that when blood-hepcidin measurements become available, this would provide a more satisfactory index of "functional iron deficiency" necessitating iron supplements. It is also theoretically conceivable that some day inhibitors of hepcidin will facilitate restoration of hemoglobin concentrations in the anemia of chronic disease by augmenting intestinal iron absorption and particularly by releasing sequestered iron from the RES for erythropoiesis.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
HOME
CURRENT ISSUE
ARCHIVES
JASN Express
ONLINE SUBMISSION
AUTHOR INFO
EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP |
