March 31st, 2014 | ASRMN Romania
Categories: There is no tags
Comments: No Comments
The chronic kidney disease in diabetes mellitus patients – an update
assoc. prof. Cristian Serafinceanu
The chronic kidney disease (CKD) associated with diabetes mellitus (DM), mainly with type 2, is still the leading cause of end stage renal disease (ESRD) in USA and Western Europe and an increasing trend is registered worldwide. Type 2 DM and arterial hypertension separately or together are accounting for about 75% of the new ESRD patients initiated on renal replacement therapies in 2011 in USA. However, in the last few years a new classification for CKD has been proposed by KDIGO (1) resulting from the conclusions of some recent trials and this is enforcing in our opinion a change in the actual paradigm of the diabetic kidney disease (DKD). Some of these changes will be examined in the next pages.
2. THE NEW PATTERNS OF THE CHRONIC KIDNEY DISEASE EVOLUTION IN PATIENTS WITH DIABETES MELLITUS
2.1. The histopathologic heterogeneity
The most striking in this view seem to be the changes in the evolution pathway of DKD; recent studies (2, 3, 4) prove that the cardiovascular morbidity/mortality risk of the patients with diabetes and any type of renal involvement is significantly greater than their risk to progress to end stage renal diasease. This high mortality risk level is related to micro- and macrovascular diabetic complications also, so although microvascular outcomes are considerably improved by reaching the BP and HbA1c targets the patients may still then be at increased risk of macrovascular events.
With the widespread use of renin-angiotensin-system (RAS) medications (ACE inhibitors and ATR1 blockers) in the last two decades and the new means for tight blood pressure and glycemic control and complex monitoring (CGM systems etc), a considerable heterogeneity of the renal pathology has been revealed, especially in type 2 DM patients without proliferative retinopathy (5,6).
Classically, the inexorable pathway of the diabetic kidney disease, first described in the early 1980s (7) has started with the glomerular hyperfiltration stage as a result of hypertrophied kidneys and glomeruls (¨nephromegalia¨); the onset of microalbuminuria, a persistent increase in the urinary albumin excretion ranging between 30-299 mg/day or 30-299 mg/g of urinary creatinin, certified the diagnosis of established diabetic kidney disease, the stage called ¨incipient¨ nephropathy.
About one-third of the patients with type 2 DM and 40% with type 1 DM were considered prone to progress beyond this point with worsening of albuminuria, clasically named as ¨overt¨ nephropathy. Decisive histopathologic changes were thought to occur in this stage of evolution, such as the nodular glomerular intercapilar sclerosis (Kimmelstiel-Wilson, KW), resulting in a very high risk for progression to ESRD. This evolution was described in patients with type 1 DM, but the diabetic CKD associated with type 2 DM has been considered to follow approximate the same pattern.
The last two decades have brought a lot of evidence that challenges these data. There is an overlap in the histopathologic and even clinic manifestations of the DKD in type 1 and type 2 DM; nevertheless, the retardation or even arrest of the DKD evolution in a large number of type 2 DM patients by using modern means of treatment and monitoring allows the highlighting of multiple renal morphologic patterns, some of them improperly identified as DKD, especially in the absence of significant retinopathy. Specifically, one – third of type 2 DM patients have displayed important tubular atrophy, interstitial sclerosis and vascular lesions (medial sclerosis, intimal hyperplazia, etc), which are determining their renal prognosis and are completely different from the chronic glomerulopaty usually known as DKD.
2.2. The non proteinuric diabetic chronic kidney disease and the reversibility of albuminuria
Large prospective trials (5,8,9) have shown that 25 to 50% of type 2 DM patients with decreased eGFR < 60 ml/min are normoalbuminuric. Overall, clinical proteinuria (¨macroalbuminuria¨) prevalence tends to wane (from 83% to 63%) in type 2 DM patients (10); instead, the frequency of normoalbuminuria associated with decreased eGFR has doubled and that of microalbuminuria tripled in the last 20 years in the same patients. In the same time, the prevalence of decreased eGFR in these type 2 DM patients remains unchanged, stregthening the opinion that between the urinary albumin excretion rate (UAER) and the decline rate of the renal function the relationship is only weak and indirect (11). In this view, microalbuminuria appears to be mainly an early marker of cardiovascular disease (endothelial disfunction), as an old hypothesis (¨Steno hypothesis¨) has stated.
A recent analysis of the NHANES Trial between 2001-2008 (12), including 18,545 patients (2798 with DM) has shown that the overall mean prevalence of normoalbuminuric CKD (NA-CKD) in the US population was 9.7% in DM patients compared to 4.3% in non diabetic individuals. Diabetic NA-CKD was less prevalent in men (OR: 0.58; 95%CI 0.39 – 0.87), less frequent also in patients with poor glycemic control (OR: 0.48; 95%CI 0.31 – 0.74) and higher blood pressure (OR: 0.25; 95%CI 0.13 – 0.50) and more frequent in white patients compared to black (OR: 0.44; 95%CI 0.29 – 0.68). The results for non-diabetic patients were similar.
Some studies (13,14) have demonstrated that abnormal UAER levels (microalbuminuria and even macroalbuminuria) are capable of regression in both type 1 and type 2 DM patients and the regression is associated with reduction in risk for ESRD and cardiovascular mortality. One important study (15) challenged relatively recent the strong thesis that glomerular sclerosis is an irreversible process, finally leading to glomeruli loss and ESRD in type 1 DM patients with DKD. In this paper the authors have shown that after ten years of normoglycemia (patients with pancreas transplantation) the mesangial expansion and even some sclerotic KW nodules have dissappeared; the process was called ¨glomerular remodelling¨.
In conclusion, NA-CKD appears to be more frequent in patients with DM, women, non – hispanic whites and in the setting of well controlled blood pressure and glycemia. These findings are supporting the new paradigm, which defines the CKD associated with DM as a dynamic and heterogenous entity whose evolution and prognosis is substantially influenced by the metabolic (glycemic) and blood pressure long term control.
3.THE THERAPEUTIC MANAGEMENT: KEY POINTS UPDATE (TABLE 1)
3.1 Blood pressure targets and management
The JNC 7 Report, published in 2003 (16) recommends a target blood pressure of 130/80 mmHg in patients with DM, as an important issue of the cardiovascular relative risk reduction. Nevertheless, in the large ACCORD Trial no benefit with regard to the composite cardiovascular endpoint (non-fatal miocardial infarction, non fatal stroke and cardiovascular death) was found after the randomization of systolic blood pressure to less than 120 mmHg versus less than 140 mmHg (17).
Another large prospective RCT, ADVANCE randomized more than 11000 patients with diabetes on an ACEI/diuretic combined pill, without regard to their BP baseline status (18); a significant reduction in DKD evolution was demonstrated (HR=0.70; 95%CI 0.56-0.88), which was entirely based on the antialbuminuric effect of the intervention. An important finding also is that there was no lower BP level at which the benefit of lowering UAER was lost.
The majority of the epidemiologic studies in the last years have clerly demonstrated the negative effect of hypertension on the evolution of the DKD; though, only few randomized trials specifically adressed BP targets for the primary or secondary prevention of DKD (17, 19, 20). In the ABCD Trial (20), the BP in the ¨normotensive¨ arm (type 2 DM patients) has been lowered to 128/75 mmHg versus 137/81 mmHg in the control group; although no significant difference in the creatinine clearance between the two groups was detected (primary endpoint), there was a reduction in transition rates from micro- to macroalbumiuria.
The last ADA Recommedations 2014 have stated that a systolic BP target of 140 mmHg should be appropriate for the majority of DM patients, but ¨ lower systolic targets (such as 130 mmHg) may be appropriate for certain individuals such as younger patients, if it can be achieved without undue treatment burden.¨.
In conclusion, even the systolic BP reduction to less than 120 mmHgdoes not show any benefit (cardiovascular or mortality) it reduces the progression of albuminuria and possible the GFR decline; if such a target is utilized in selected patients, a close monitoring is mandatory. After all the recent studies (Table 2) comparing the BP targets of 130/80 mmHg and 140/85 mmHg, there are no definitive evidences looking at renal outcomes.
3.2 Glycemic control
Large controlled randomized prospective trials (Table 3) have shown in the last 20 years the major effect of intensive glycemic control on the development and outcomes of microvascular complications in both type 1 and type 2 DM patients.
The first of these trials was DCCT (21). In this trial, after a median follow up of 6.5 years of patients with recent type 1 DM and a mean difference of 1.9 % between the intensively treated versus conventionally treated, a 54% relative risk reduction dor the development of new onset macroalbuminuria was registered. The DCCT participants were further followed up for an additional 8 years period in the EDIC (22) Study (Epidemiology of the Diabetes Interventions and Complications Study). Despite the evolution of mean HbA1c of the two groups, which converged to a value of approximately 8% after 1 year, previous intensive glycemic control afforded long term benefits: the relative risk for macroalbuminuria ad serum creatinine elevation of > 2mg/dl was 0.8 (95%CI 0.7-0.9) in the conventionally treated versus 0.3 (95%CI 0.2-0.4) in intensively treated patients; this result allowed the identification of the ¨metabolic memory¨ in type 1 DM patients.
The prospective RCTs in type 2 DM patients provided supplemmentary evidence of the benefit of intensive glycemic control on the incidence and evolution of the diabetic chronic kidney disease. In the UK Prospective Diabetes Study (23) (UKPDS) a mean HbA1c reduction of only 0.9% between the intensively and conventionly treated patients (7% versus 7.9%) provided a 25% relative risk reduction in microvascular outcomes and a relative risk of doubling the serum creatinine of 0.5 (95%CI 0.3-0.7).
Following this landmark study, three large RCT trials in type 2 DM patients recently published (ACCORD, ADVANCE and VADT) further substantiate a significant benefit in preventing/delaying macroalbuminuria onset and progression in patients with long term type 2 DM and high cardiovascular risk.
Nevertheless, all the cited studies like the RCTs examining different BP targets relationships with diabetic CKD also, failed to demonstrate a clear benefit on clinical endpoints for diabetic CKD. This discordance between UAER endpoints versus clinical renal endpoints in DM patients still undermines the results of the RAAS inhibition as well as glycemic interventions studies and the potential indication for primary prevention of the diabetic CKD, even in selected DM patients.
A recent study (24) had evidenced the importance of the new biomarker of the glycemic control glycated albumin (GA). The glycated hemoglobin levels are depending not only of the average blood glucose but also on the lifespan of red blood cells; so, the patients with hemoglobin disorders or anemia of any cause may have erroneous HbA1c levels. It is a fact that patients with diabetic CKDoften suffer from various types of anemia (erythropoietin and/or iron defficiency, denutrition, hemolysis etc)and consequently they are frequently treated with iv iron, erythropoietin and blood transfusions. Serum GA levels are not influenced by anemia and associated treatments, so it is a potentially useful biomarker for glycemic control in patients with diabetic CKD. GA reflects the status of glycemic control more rapidly (2 to 3 weeks) than HbA1c (2 to 3 months) and is particularly more precise in those patients with wide variations in blood glucose such as patients with diabetic CKD.
3.3 Screening and referral for diabetic CKD patients
The above changes in the paradigm of the diabetic CKD natural history and the running therapeutic key points changes indicate the importance of monitoring both the UAER and eGFR; moreover, the glomerular function should be checked by annual screening of serum creatinine and by calculating eGFR also, using either MDRD or CKD – EPI equations.
The screening for albuminuria is recommended on an annual basis also (after 5 years duration of type 1 DM and from the onset of type 2 DM), either using an urine dipstick (high rate of false results when the urine is extremely concentrate or diluted) or an albumin/creatinine ratio in a spot urine collection (most accurate when using the first morning urine specimen). The diagnosis of micro- or macroalbuminuria should be based on two out of three positive urine/creatinine ratios results (30-300 mg/g for microalbuminuria and more than 300 mg/g for macroalbuminuria, respectively) in urine specimens collected over a 3-6 months period. In cases of type 1 DM associated with metabolic syndrome screening for albuminuria from the onset of disease should be acceptable.
Periodic monitoring of UAER, once the diagnostic of albuminuria established is questionable: there are epidemiologic data showing a decreased risk of eGFR deterioration with albuminuria remission (12, 13, 25); on the other hand, as discussed above, there was detected a discordance between changes in the UAER and clinical renal endpoints. It is widely recognized that an increasing UAER is a marker of poor prognostic in diabetic CKD patients, identifying individuals that need an intensive effort for cardiovascular risk reduction.
The nephrology referral of diabetic patients is a sensitive point because of its decisive effect on the renal/vital prognosis of these patients. A good reason for early referral would be suspicion of non diabetic CKD (based on the presence of microscopic hematuria or rapid changes in UAER or eGFR). One must keep in mind that eGFR decline in diabetic CKD is highly variable and often non – linear (26), depending on glycemic and BP control and the onset of CKD specific complications (hyperphosphatemia, hyperparathyroidism, anemia and netabolic acidosis). The patients with DM whose eGFR decline rate is exceeding 10 ml/min/year should be referred imediatly for nephrologic exam.
The patients with eGFR values of less than 30 ml/min/1.73 m2 (stage 4 of CKD) should be also referred to a nephrologist to prepare them for the initiation of a renal replacement therapy (renal transplantation or dialysis). Our own data, recently published (27) have shown that late initiation of dialysis, mainly due to late referral to nephrology, is significantly associated with higher rate of early mortality after starting dialysis.
3.4. The renin-angiotensin system (RAS) blockade
Many prospective RCTs have tested in the last twenty years the hypothesis that RAS inhibitors, either inhibitors of the angiotensin-converting enzyme (ACE-I) or angiotensin receptors blockers (ARB) prevent the development or slow the progression of diabetic CKD (29-34). All analysis of these studies revealed that the blockade of RAS provided certain benefits in reducing the risk of macroalbuminuria compared to placebo but no benefit when compared with other anti hypertensive agent. The large majority of placebo controlled trials have shown significant differences of the mai blood pressure values between groups.
When the clinical end points (doubling of serum creatinine, end stage renal disease incidence and death) are examined the results are frequently conflicting (Table 4). The trials designed to select patients with high baseline risk for progression to end stage renal disease have shown a benefit from the inhibition of RAS (31, 32), especially in reducing the eGFR decline rates and macroalbuminuria development. On the other hand, large trials like ALLHAT (35) or ADVANCE (36), which were designed for primary prevention of diabetic CKD in patients with DM and high cardiovascular risk failed to reveal any benefit to reduction the risk of end stage renal disease from RAS inhibition.
Another topic which was recently re-evaluated is the dual RAS blockade, defined as the concomitent use of an ACE-I and an ARB. The dual blockade was based on data from early 1990s showing the reduction of proteinuria and prevention of microalbuminuria. Recent studies have demonstrated that microalbuminuria is not a necessary criterion for the diagnosis of diabetic CKD and the development of proteinuria is not always concurrent (even usually is highly correlated) with the progression to renal failure.
The rationale for dual blockade is that neither ACE-Is nor ARBs can supress completely the RAS (37) because of the ¨escape¨ phenomenon: alternate non ACE-dependent pathways (chymase) for angiotensinogen activation to angiotensine II. These mechanisms are probably responsible for the recurrence of the angiotensine II levels back to baseline after approximate six to nine months of ACE-I treatment.
Unfortunately, the majority of RCT testing the dual blockade in patients with diabetic CKD have used as endpoints microalbuminuria or proteinuria and not the eGFR decline rate, but the albuminuria endpoints are not recognized anymore as surrogate biomarkers for diabetic CKD progression; so, there is no current clear data about an additional benefit from a more profound RAS inhibition using both ACE-Is and ARBs.
The ONTARGET Study (38) was the only RCT using an ARB (telmisartan) associated to an ACE-I (ramipril) and having clinical end points (doubling of serum creatinine, need for dialysis or death) in a large cohort of approximate 25,000 patients with DM and early CKD. The results have shown no benefit from dual blockade in terms of CKD pregresssion (doublind of serum creatinine or end stage renal disease) or mortality; moreover, the patients in the dual blockade group had a greater risk for acute renal injury and hyperkaliemia.
In conclusion, although the available data have demonstrated the ability of dual RAS blockade to lower albuminuria with an additional 20 to 40 % compared to monotherapy and independently of the blood pressure reduction, no additional reduction in the cardiovascular risk or even in CKD progression was revealed. Only the patients with advanced proteinuric diabetic CKD could eventually have an indiaction for dual blockade, but the serious risks of hyperkaliemia and acute renal injury must be careful evaluated.
3.5 New and potential drugs for the treatment of diabetic chronic kidney disease
Some existing and other new drugs are being evaluated for their possible ability to interrupt the disease process of diabetic CKD. In a recent meta-analysis on this topic (39) fifteen compounds and 24 studies have been reviewed, ten compounds having evidence of beneficial effects in patients with diabetic CKD: improvements in eGFR, urinary albumin-to-creatinine ratio, proteinuria and serum creatinine.
Antifibrotic agents (pirfenidone, doxycicline and aliskiren). Pirfenidone is exertind its antifibrotic action through inhibition of TGF-β in animal models; in human subjects significant increse in eGFR was noted after dose of 1200mg/d compared to palcebo after one year. The antiproteinuric effect of doxycicline, initially significantly more important than placebo dissapeared after 6 months. Aliskiren is a direct renin inhibitor which also exhibits antifibrotic properties exerted through inhibition of expression of TGF-β. The recent trial ALTITUDE (40) tried to evaluate this effect in human subjects but the study has been stoped because of increased risk for hypotension, hyperkaliemia and non fatal stroke after 2 years.
Vitamins of the group B and derivatives (pyridoxamine and benfotiamine). Pyridoxamine is a derivative from B vitamins group and acts as an AGE inhibitor. Benfotiamine, another derivative has shown beneficial antiproteinuric effects in animal studies but this effect was not confirmed in human subjects. A prospective RCT testing B vitamins therapy (B6 plus B12 and folic acid) in 238 DM (41) patients demonstrated a sgnificant reduction in eGFR after 3 years and a 2-fold increase in stroke, niocardial infarction and all cause mortality in the B vitamin group compared to placebo (HR: 2.0; 95%CI1.0-4.0; p=0.04).
Antioxidants modulators of inflammation (bardoxolone and pentoxifylline) Bardoxolone resembles to prostaglandines and is classified as an antioxidant with properties of inflammation modulator. A recent RCT, BEAM, was designed to determine effects of bardoxolone on eGFR in patients with type 2 DM and diabetic CKD (baseline eGFR between 25-40 ml/min/1.73 m2) versus placebo at 52 weeks. The results demonstrated a significant dose dependent increase of eGFR in bardoxolone groups.
Another RCT, BEACON, was designed to to examine the effects of bardoxolone on progression of diabetic CKD to end stage renal disease and cardiovascular mortality in type 2 DM patients; unfortunately, the study was stoped in october 2012 because of significant increase of cardioavscular events in the treatment groups.
Pentoxifylline is a methyl xantine derivative and was long time used for improvement in peripheral blood flow through reduction of blood viscosity in patients with peripheral artery disease (intermitent claudication).There were some small studies on human patients showing that the weak antiproteinuric effect of pentoxifylline could be useful in association with an ACE-I. An ongoing RCT named PREDIAN will measure the difference in eGFR decline rate between a control group and a pentoxifylline test group (42).
Glycosaminoglycans (sulodexide) Sulodexide is classified as glycosaminoglycan, containing 80% low molwcular weight heparin and 20% dermatan sulphate. The DiN.A.S. Study (RCT completed in 2002) have demonstrated the antiproteinuric properties of sulodexide in diabetic patients with CKD. These results were later confirmed in other two trials in patients with diabetic CKD.