REVIEW
`EARLY DETECTION AND MANAGEMENT OF DIABETIC KIDNEY DISEASE
`
`Rationale and Strategies for Early Detection and Management
`of Diabetic Kidney Disease
`
`BRIAN RADBILL, MD; BARBARA MURPHY, MD; AND DEREK LEROITH, MD, PHD
`
`Diabetic kidney disease (DKD) occurs in 20% to 40% of patients
`with diabetes mellitus and is the leading cause of chronic kidney
`disease and end-stage renal disease in the United States. Despite
`the American Diabetes Association and the National Kidney Foun-
`dation advocating annual screening of diabetic patients, DKD
`remains underdiagnosed in the diabetic population. Early recogni-
`tion of diabetic nephropathy by health care professionals is vital
`for proper management. The presence of microalbuminuria is
`particularly important as even low levels of dipstick-negative
`albuminuria indicate early disease long before a diminished glo-
`merular filtration rate and are associated with an elevated cardio-
`vascular disease risk. Like all forms of chronic kidney disease,
`DKD causes a progressive decline in renal function that, despite
`current treatment strategies, is largely irreversible. Many patients
`with DKD might be expected to develop end-stage renal disease,
`but many more patients will likely die of a cardiovascular event
`before renal replacement therapy is needed. Therefore, a renewed
`focus on cardiovascular risk factor reduction and a timely nephrol-
`ogy consultation with an emphasis on patient education is essen-
`tial to proper DKD management.
`Mayo Clin Proc. 2008;83(12):1373-1381
`
`ACE = angiotensin-converting enzyme; ACR = albumin-to-creatinine
`ratio; ADA = American Diabetes Association; ARB = angiotensin II
`receptor blocker; CKD = chronic kidney disease; CVD = cardiovascular
`disease; DKD = diabetic kidney disease; DM = diabetes mellitus; ESRD =
`end-stage renal disease; GFR = glomerular filtration rate; HbA1c =
`glycosylated hemoglobin; MDRD = Modification of Diet in Renal Disease;
`NKF = National Kidney Foundation; RAAS = renin-angiotensin-aldoste-
`rone system
`
`The prevalence of both diabetes mellitus (DM) and
`
`chronic kidney disease (CKD) is steadily increasing in
`the United States. Current estimates suggest that 7% of the
`population (approximately 21 million people) have DM
`and that 13% of the population (approximately 26 million
`people) have CKD.1,2 It may be argued that histologic
`findings of diabetic nephropathy, including glomerular
`basement membrane thickening and mesangial matrix ex-
`pansion, are present in all patients with DM. However,
`diabetic kidney disease (DKD), defined as an elevated
`albumin excretion rate in a person with DM, occurs in 20%
`to 40% of patients with DM and is the leading cause of CKD
`and end-stage renal disease (ESRD) in the United States.1,3
`The increased prevalence of CKD is no doubt linked to the
`increased prevalence of DKD and DM, which is attributed
`largely to a dramatic increase in the obesity rate.4
`In response to the growing prevalence of DKD and DM,
`which is increasingly recognized as an epidemic, the
`American Diabetes Association (ADA) and the National
`Kidney Foundation (NKF) have advocated annual screen-
`
`ing for DKD in patients with DM by measuring their serum
`creatinine and albuminuria levels.1,3 Despite these recom-
`mendations, DKD remains underdiagnosed in the DM popu-
`lation.5-7 In a review of Medicare beneficiaries’ records,
`proteinuria was measured in only 63% of patients with DM.6
`Furthermore, in a survey of more than 1000 primary care
`physicians, only 12% detected microalbuminuria in more
`than half of their patients with type 2 DM.7
`Assessment of microalbuminuria is particularly impor-
`tant in diagnosing DKD because low levels of dipstick-
`negative albuminuria are an early clinical manifestation of
`diabetic nephropathy that may present several years before
`development of a diminished glomerular filtration rate
`(GFR). Spot urine samples have replaced the need for
`timed urine collections and can be used to easily identify
`patients with elevated albumin excretion rates by measur-
`ing the albumin-to-creatinine ratio (ACR). Once an el-
`evated ACR has been detected, interventions should be
`initiated to slow the progression of DKD and possibly
`minimize the increased cardiovascular risk associated with
`DKD, a risk that exists even in the early stages of DKD.
`Although DM has long been identified as a cardiovas-
`cular disease (CVD) risk equivalent, only recently has
`CKD been more widely recognized by primary care physi-
`cians in the United States as an independent risk factor for
`CVD and all-cause mortality.8-11 In a study of more than 1
`million ambulatory adult patients, the risk of a cardiovas-
`cular event and death due to any cause increased at every
`level of CKD below a GFR of 60 mL/min per 1.73 m2, with
`a nearly 3.5-fold increased risk of a cardiovascular event
`and a 6-fold increased risk of death for those with a GFR of
`less than 15 mL/min per 1.73 m2 (ie, CKD stage 5).11
`Furthermore, microalbuminuria alone has been associated
`with an increased risk of cardiovascular disease, both in
`patients with and without DM.12-14 Therefore, in patients
`with DKD, the cardiovascular risks of DM and CKD are
`additive and increase as the kidney disease progresses.15-17
`
`From the Division of Nephrology (B.R., B.M.) and Division of Endocrinology,
`Diabetes, and Bone Disease (D.L.), Mount Sinai School of Medicine, New
`York, NY.
`
`Individual reprints of this article are not available. Address correspondence to
`Derek LeRoith, MD, PhD, Mount Sinai School of Medicine, Department of
`Medicine, One Gustave L. Levy Pl, Box 1055, Atran Bldg 4-36, New York, NY
`10029-6574 (derek.leroith@mssm.edu).
`© 2008 Mayo Foundation for Medical Education and Research
`
`Mayo Clin Proc. • December 2008;83(12):1373-1381 • www.mayoclinicproceedings.com
`
`1373
`
`MPI EXHIBIT 1142 PAGE 1
`
`

`

`EARLY DETECTION AND MANAGEMENT OF DIABETIC KIDNEY DISEASE
`
`TABLE 1. Stages of Chronic Kidney Disease and Recommended Treatmenta
`GFR
`(mL/min per 1.73 m2)
`(cid:42)90
`
`Treatmentb
`Manage comorbid conditions, slow progression,d reduce CVD risk
`
`Stage
`1
`
`2
`
`3
`
`Description
`Kidney damagec with normal or
`elevated GFR
`Kidney damagec with mildly
`reduced GFR
`Moderately reduced GFR
`
`60-89
`
`30-59
`
`Estimate progression as follows: compare serial estimated GFRs using serum
`creatinine and MDRD calculation, track ACR
`Evaluate and manage complications as follows: (1) measure serum
`phosphorus level, consider use of phosphate binders and low-phosphorus diet;
`(2) measure vitamin D and parathyroid hormone levels, consider use of
`vitamin D supplementation; (3) measure hemoglobin, consider use of ESA
`Prepare for kidney replacement therapy
`15-29
`Severely reduced GFR
`4
`Kidney replacement (if uremia present)
`<15 or dialysis
`Kidney failure (ESRD)
`5
`a ACR = albumin-to-creatinine ratio; CVD = cardiovascular disease; ESA = erythropoietic stimulating agent; ESRD = end-stage renal disease; GFR =
`glomerular filtration rate; MDRD = Modification of Diet in Renal Disease.
`b Includes treatments from preceding stages.
`c Defined as abnormalities on pathologic, urine, blood, or imaging tests.
`d Glycemic control plus angiotensin-converting enzyme inhibitor or angiotensin II receptor blocker.
`Data from Ann Intern Med.9
`
`The current article presents results of a literature review
`conducted to clarify the rationale and strategies for early
`detection and management of DKD.
`
`METHODS
`The National Library of Medicine’s PubMed database was
`used to conduct a review of literature published between
`January 1976 and June 2008. The following key terms were
`used in the search: diabetes, kidney disease, microalbumin-
`uria, glomerular filtration rate, and diabetic nephropathy.
`
`RESULTS
`
`OVERVIEW OF RENAL PATHOPHYSIOLOGY
`The kidneys receive 25% of the cardiac output of blood.
`Although 20% of renal plasma flow (ie, approximately 180
`L) is filtered through the glomerulus, only small amounts
`of protein can be detected in the urine.18 Several plasma
`proteins are freely filtered, whereas others are prevented
`from crossing the glomerular filtration barrier, based on the
`proteins’ molecular size and charge. The existence of
`several restrictive pores and of a glomerular charge barrier
`has been proposed to explain why the glomerulus is rela-
`tively impermeable to proteins of greater molecular weight
`(ie, >100 kDa) and to negatively charged proteins (eg,
`albumin).18
`More recently, it has been suggested that, under normal
`conditions, a substantial amount of plasma protein, possi-
`bly at nephrotic levels, is filtered through the glomerulus,
`but proteinuria is prevented because of proximal tubule
`cell retrieval.19 According to this idea, damage that disrupts
`the glomerular filtration barrier, or possibly the proximal
`tubular system, allows larger, negatively charged proteins
`that are normally contained within the serum to pass into
`
`the urine. The presence of such proteins, typically albumin,
`in the urine is an abnormal condition and is often one of the
`first signs of various forms of CKD, including DKD.
`Chronic kidney disease is defined as kidney damage
`identified by proteinuria or by a GFR of less than 60 mL/
`min per 1.73 m2 body surface area (with or without evi-
`dence of kidney damage) for 3 months or longer.8,9 Table 1
`shows the stages of CKD and the recommended treatments
`at each stage. In patients with DKD, the disease process
`begins with renal hypertrophy and hyperfiltration resulting
`from elevated renal plasma flow. In patients with type 1
`DM and type 2 DM, hyperglycemia leads to increases in
`GFR of approximately 5% to 10%.20-23 Although the
`mechanism is not completely understood, a correlation
`exists between glycosylated hemoglobin (HbA1c) and GFR,
`and normalization of blood sugar levels has been shown
`to normalize GFR.24,25 Other factors that influence hyper-
`filtration include increased ketone concentration, increased
`activity of the growth hormone/insulin-like growth factor
`system,26 and disturbances in renal prostaglandins and
`the kallikrein-kinin system. In early-stage CKD, these
`abnormalities are frequently associated with enlarged
`kidneys.27
`Hyperfiltration is typically followed by the loss of the
`negatively charged glomerular filtration barrier, allowing for
`negatively charged proteins, such as albumin, to pass
`through the glomerulus and into the urinary space. The
`presence of these proteins in the urinary space elevates uri-
`nary albumin excretion and produces microalbuminuria.27
`Microalbuminuria is defined as an albumin excretion rate
`between 30 and 300 mg per 24 hours, a range higher than
`the normal rate (<30 mg per 24 hours) but below the rate
`detectable by the standard urine dipstick method.8 Over-
`excretion of albumin typically increases at a rate of 15%
`per year28 and can result in macroalbuminuria (>300 mg per
`
`1374
`
`Mayo Clin Proc. • December 2008;83(12):1373-1381 • www.mayoclinicproceedings.com
`
`MPI EXHIBIT 1142 PAGE 2
`
`

`

`24 hours) or even nephrotic-range proteinuria (>3.5 g per
`24 hours).
`In general, once macroalbuminuria (frank proteinuria)
`sets in, GFR begins to decline.29 Progressive mesangial and
`interstitial capillary occlusion then occur, restricting the
`glomerular filtration surface and leading to a further de-
`crease in GFR. Some proteins are reabsorbed by the renal
`tubules and accumulate in tubular epithelial cells. This
`accumulation induces the release of vasoactive and in-
`flammatory cytokines, which damage the renal tubules
`and lead to tubular atrophy and interstitial fibrosis.30 A
`negative feedback loop is thereby initiated, wherein in-
`creased proteinuria leads to increased tubulointerstitial
`injury and renal scarring, both of which further reduce
`GFR.30
`Both hypertension and hyperglycemia are important in
`the development and progression of microalbuminuria and
`DKD. Table 2 presents a list of disorders associated with
`microalbuminuria. Several studies have shown that blood
`pressure elevations either precede or occur in conjunction
`with microalbuminuria in patients with both type 1 DM and
`type 2 DM.29,31,32 Among patients with type 1 DM and DKD,
`those with increased urinary albumin excretion were found
`to be prehypertensive (120-139/80-89 mm Hg) at baseline,
`and their blood pressure and albuminuria levels increased in
`synch thereafter.31,32 These elevations happened even though
`overt hypertension was not present before the onset of
`microalbuminuria. In patients with type 1 DM and DKD,
`blood pressure elevations before the onset of DM correlated
`with the future development of microalbuminuria.33
`As previously mentioned, hyperglycemia can affect
`GFR and is necessary for the development of DKD.
`Likely mechanisms by which elevated glucose levels
`cause kidney damage include accumulation of advanced
`glycation end products, glucose-induced growth factor
`expression, and increased expression of inflammatory
`factors. However, hyperglycemia alone is insufficient to
`cause renal dysfunction.30
`Most patients with DM never have clinically evident
`DKD, despite poor glycemic control. The absence of DKD
`in these patients suggests a genetic predisposition for
`DKD. The existence of such a predisposition is supported
`by studies showing an increased risk of nephropathy
`among people with a family history of the disorder.34-36
`Nevertheless, in susceptible individuals, hyperglycemia
`plays a crucial role in the progression of DKD from
`microalbuminuria to renal insufficiency and ESRD, as
`shown in type 1 DM by the Diabetes Control and Compli-
`cations Trial37 and in type 2 DM by the United Kingdom
`Prospective Diabetes Study.38 Both these studies conclu-
`sively showed that the development and progression of
`DKD are strongly correlated with deficiencies in glucose
`
`EARLY DETECTION AND MANAGEMENT OF DIABETIC KIDNEY DISEASE
`
`TABLE 2. Disorders Associated With Microalbuminuria
`Elevated blood pressure
`Dyslipidemia
`Elevated fibrinogen and plasminogen activator inhibitor 1
`Increased insulin resistance
`Increased sodium disorders and related disorders
`Increased transcapillary escape rate of albumin
`Impaired basal endothelium-dependent vasorelaxation
`Increased left ventricular volume
`Diabetic retinopathy
`Diabetic neuropathy
`Peripheral vascular disease
`Silent ischemic heart disease
`
`control, verifying that glycemic control remains one of the
`cornerstones of treatment of DKD.
`
`SCREENING AND MONITORING TECHNIQUES
`The ADA recommends that both microalbuminuria and
`serum creatinine levels be assessed annually in patients
`with DM to screen for DKD.3 For patients with type 1 DM,
`screening should begin 5 years after diagnosis because it
`takes at least that long for signs of nephropathy to develop.
`For patients with type 2 DM, screening should begin imme-
`diately at diagnosis because the precise onset of DM is
`often less clear, and the kidneys may have already sus-
`tained damage from years of undiagnosed hyperglycemia
`and/or hypertension.
`After evidence of DKD has been detected, ongoing
`evaluations should be based on measurements of GFR.3
`However, in clinical practice, albuminuria is also typically
`measured to monitor disease progression and optimize
`therapy.
`The following sections review the various methods used
`to measure GFR and albuminuria, focusing on the benefits
`and limitations of each.
`Glomerular Filtration Rate. An index of functioning
`renal mass, GFR assessment is the most reliable method of
`detecting and monitoring renal impairment. Glomerular
`filtration rate can be measured directly or it can be esti-
`mated indirectly using the Modification of Diet in Renal
`Disease (MDRD) or Cockcroft-Gault equations. Simple
`measurement of serum creatinine is not recommended as
`an estimate of GFR because creatinine levels are greatly
`influenced by an individual’s muscle mass, and thus simple
`measurements may overestimate or underestimate true
`GFR. Another reason that serum creatinine measurements
`may lead to an overestimation of GFR is that creatinine is
`cleared via secretion by the proximal tubule, and extrarenal
`excretion of creatinine is common in patients with more
`advanced CKD.3,8,9
`Direct measurement of the fractional excretion of inulin,
`a fructose polysaccharide, is considered the criterion stan-
`dard for GFR measurement. Inulin is inert, freely filtered at
`
`Mayo Clin Proc. • December 2008;83(12):1373-1381 • www.mayoclinicproceedings.com
`
`1375
`
`MPI EXHIBIT 1142 PAGE 3
`
`

`

`EARLY DETECTION AND MANAGEMENT OF DIABETIC KIDNEY DISEASE
`
`the glomerulus, and neither secreted, reabsorbed, synthe-
`sized, nor metabolized by the kidneys. However, using
`inulin infusion to measure GFR is expensive, cumbersome,
`and not widely available. An alternative method for mea-
`suring GFR involves a single injection of a radioisotopic
`filtration marker, such as technetium Tc 99m DTPA
`(diethylenetriaminepentaacetic acid) or iothalamate I 125.
`This approach provides an accurate measure of GFR in
`cases of renal insufficiency, but it can overestimate GFR in
`healthy individuals and is also not widely available.8,39
`The ADA and NKF recommend measuring the serum
`creatinine level and then using that value in either the
`MDRD or Cockcroft-Gault equations to estimate GFR.1,3
`Both these equations take into account variations in creati-
`nine across age and sex, and the MDRD calculation also
`takes ethnicity into account.40,41 The widely used MDRD
`calculation is considered more accurate than the Cockcroft-
`Gault equation for patients with CKD stage 2 or greater
`(GFR <90 mL/min per 1.73 m2).9 The MDRD equation was
`developed on the basis of direct GFR measurements and
`clearance of iothalamate 125I in a study of 1628 patients of
`various ethnicities who had a variety of kidney disorders
`(6% had DM).41 The MDRD was then validated in another
`group, consisting of more than 500 individuals.41
`In general accuracy studies, more than 90% of GFR
`values estimated with the MDRD equation were within
`30% of directly measured creatinine values, compared with
`75% of values estimated with the Cockcroft-Gault equa-
`tion.8 Accuracy of estimates is improved if the clinical
`laboratory calibrates the creatinine measurement to the
`Cleveland Clinic’s database, which includes approxi-
`mately 9000 GFR measurements.42 For this reason, many
`clinical laboratories are now undergoing the necessary
`steps to calibrate creatinine measurement.
`Despite ADA and NKF recommendations, neither the
`MDRD calculation nor the Cockcroft-Gault equation has
`been validated for use in cases of diabetic nephropathy.9 A
`recent accuracy study of patients with DM and micro-
`albuminuria found that, although both the MDRD and
`Cockcroft-Gault equations correlated with directly mea-
`sured GFR, both equations significantly underestimated the
`filtration rate, especially in patients with microalbumin-
`uria.43 The rate of renal decline was also significantly under-
`estimated. The sensitivity of the equations to detect renal
`impairment was 72% for MDRD and 66% for Cockcroft-
`Gault. Furthermore, the use of these calculations led to accu-
`rate identification of CKD (as confirmed by a measured GFR
`<60 mL/min per 1.73 m2) in only 51% (MDRD) and 66%
`(Cockcroft-Gault) of study participants.43
`In a study of 169 patients with type 2 DM and
`macroalbuminuria, both equations underestimated GFR,
`although MDRD performed better than Cockcroft-Gault.44
`
`One study evaluated the equations by repeatedly measuring
`GFR with iothalamate for 10 years in 87 patients with type
`2 DM and varying degrees of renal function: hyperfiltration,
`normal renal function, and CKD stage 2 or 3.45 Both the
`MDRD and Cockcroft-Gault equations significantly under-
`estimated GFR in patients with hyperfiltration and normal
`renal function. Nevertheless, in patients with CKD stage 2 or
`stage 3, GFR estimates made with MDRD closely matched
`iothalamate-determined GFR.45
`The reason that the accuracy of the MDRD and
`Cockcroft-Gault equations is diminished in cases of DM is
`unknown. Creatinine clearance rate varies with age, sex,
`ethnicity, and body weight, and it is also affected by
`extremes of muscle mass and dietary intake. The NKF
`recommends that GFR be measured using direct clearance
`methods in patients with severe obesity, a population that
`includes many patients with type 2 DM but few with type 1
`DM.1 In patients with mild renal impairment (ie, CKD
`stage 1 or 2), the ability of the equations to estimate GFR is
`hampered by hypertrophy and hyperfiltration, which com-
`pensate for damaged nephrons8 and may account for some
`of the observed inaccuracies.
`An alternative approach being investigated is the mea-
`surement of cystatin C concentration as a surrogate for
`GFR. Cystatin C is a plasma protein that is freely filtered
`through the glomerulus and almost completely reabsorbed
`and catabolized by tubular cells. Several recent studies
`have examined the use of cystatin C concentration as an
`alternative method of estimating GFR. However, cystatin
`C is not yet used clinically because it is not widely avail-
`able and is not currently recommended by either the ADA
`or the NKF.
`Preliminary results suggest that cystatin C measure-
`ments may more accurately predict GFR than the MDRD
`or Cockcroft-Gault equations in patients with DM. In one
`study of 52 white patients with type 2 DM, the diagnostic
`accuracy of cystatin C measurements was 90% for iden-
`tifying GFR at rates of less than 80 mL/min per 1.73 m2,
`significantly greater than serum creatinine measurements
`alone (77%) or estimates made with the Cockcroft-Gault
`equation (85%).46 A 4-year follow-up study of 30 Pima
`Indians with type 2 DM showed that GFR estimates based
`on cystatin C were numerically similar to GFR, as deter-
`mined by iothalamate clearance, and that declining trends
`in renal function were correlated between the 2 measures
`(r=0.77).47 By contrast, GFR estimates made with the
`MDRD or Cockcroft-Gault equation did not correlate well
`with iothalamate clearance (r<0.35).47
`These provocative results await confirmation by larger
`studies. If the results are confirmed, cystatin C measure-
`ments may be used to arrive at more accurate assessments
`of CKD stage.
`
`1376
`
`Mayo Clin Proc. • December 2008;83(12):1373-1381 • www.mayoclinicproceedings.com
`
`MPI EXHIBIT 1142 PAGE 4
`
`

`

`Albuminuria. Albuminuria can be assessed by timed
`collections, both overnight and 24-hour collections, and by
`spot urine tests used to measure ACR. Urine dipstick tests
`alone are not recommended for patients with DM because
`urinary protein levels vary with hydration and other fac-
`tors, potentially leading to false-positive or false-negative
`results.3
`The 24-hour timed collection of albuminuria remains
`the preferred method for the quantitative assessment of
`proteinuria; however, it is inconvenient, and over collec-
`tion or under collection errors frequently result from
`missed or improperly timed samples. Overnight timed col-
`lections represent an alternative measure, but the shorter
`collection interval makes the sensitivity of overnight tests
`particularly vulnerable to under collection.8
`The ADA and NKF now recommend measurement of
`ACR with a spot urine test to screen for diabetic nephropa-
`thy.3 Several studies have shown clinical equivalency of
`ACR and 24-hour collections.8,48-50 Both albumin and cre-
`atinine are highly soluble, and their dilution in urine is
`similar. Because creatinine excretion is generally constant,
`the ratio of albumin to creatinine accurately represents
`protein excretion during a 24-hour period.8
`Several factors can increase urinary albumin over
`baseline values, leading to false-positive results, even
`when ACR is used as a measure. These factors include
`exercise within 24 hours of the urine test, urinary tract
`infection, fever, heart failure, marked hyperglycemia,
`marked hypertension, and protein intake. Furthermore, uri-
`nary albumin excretion has a notable intraindividual coeffi-
`cient of variation, possibly as high as 40%.51 To minimize
`this variability, first-morning-void urine samples are rec-
`ommended. However, tests with positive results should be
`repeated, and a patient should not be considered to have
`elevated urinary albuminuria until 2 of 3 abnormal results
`have been obtained within a 3-month to 6-month time
`frame.1,3
`Of note, a high-normal baseline level of albuminuria or
`a substantial increase in the level of albuminuria, even if
`still within the reference range, may signify future develop-
`ment of DKD.52 For this reason, tests having such border-
`line negative results may require closer (eg, 6-month) fol-
`low-up, especially for patients at increased risk of DKD.
`For patients with documented renal impairment, annual
`evaluations of ACR should continue to assess disease pro-
`gression and to monitor response to therapy (Table 1).
`
`RATIONALE FOR EARLY SCREENING
`Slowing Progression to ESRD. Diabetic kidney dis-
`ease, like all forms of CKD, causes a progressive decline in
`renal function that may be retarded via several treatment
`strategies. Early recognition of DKD allows clinicians to
`
`EARLY DETECTION AND MANAGEMENT OF DIABETIC KIDNEY DISEASE
`
`optimize medical management and to educate patients
`about CKD so that patients can take measures to preserve
`residual renal function. Such measures may include weight
`loss, a low-protein diet, smoking cessation, and nephro-
`toxin avoidance.53 In particular, use of nonsteroidal anti-
`inflammatory drugs should be discouraged.
`Furthermore, early awareness of DKD may prompt cli-
`nicians to adjust dosages of antidiabetes agents and to
`consider more frequent diagnostic tests. Metformin hydro-
`chloride, a widely prescribed antidiabetes agent, may gen-
`erate lactic acidosis in patients with an estimated GFR of
`less than 60 mL/min per 1.73 m2 and should be discontin-
`ued when the patient’s serum creatinine level increases
`higher than 1.4 mg/dL in women and 1.5 mg/dL in men.54
`In addition, use of intravenous contrast dye and oral so-
`dium phosphate solutions may precipitate contrast-induced
`nephropathy or acute phosphate nephropathy, respectively,
`in patients with impaired renal function.
`Patient Preparation in Cases of ESRD. Despite ag-
`gressive measures, ESRD may be expected to develop in
`many patients with DKD, and ultimately some form of
`long-term renal replacement therapy will be needed. In the
`United States, the overwhelming majority of patients with
`ESRD undergo hemodialysis,55 but preemptive living-re-
`lated or living-unrelated donor kidney transplant is often
`feasible with appropriate planning.
`Preparation for renal replacement therapy requires fo-
`cused patient education and timely referrals to a nephrolo-
`gist, vascular surgeon, and kidney transplant center. The
`success of such measures begins with early DKD recogni-
`tion by the primary care physician.
`Cardiovascular Risk Associated With CKD. Previous
`nephropathy screening guidelines for patients with DM
`focused on retarding progression to ESRD. However, in
`addition to being a risk factor for renal failure, CKD is now
`widely recognized as a major risk factor for CVD.56 In a
`retrospective claims-based study of more than 1 million
`Medicare enrollees aged 65 years and older, the risk of
`cardiovascular events was significantly increased in those
`with either CKD or DM alone, but cardiovascular risk was
`greatest when both conditions were present (Figure).57
`These results are supported by prospective epidemiological
`studies that also showed an increased risk of CVD in pa-
`tients with renal insufficiency.10,11,58,59
`Elevated cardiovascular risk occurs early in the devel-
`opment of CKD, as demonstrated by studies showing that
`even low levels of albuminuria are predictive of CVD.12-14
`Using data collected from the Heart Outcomes Prevention
`Evaluation study, investigators found that the relative risk
`of myocardial infarction, stroke, and death due to CVD in
`patients with microalbuminuria was 1.97 among those with
`DM and 1.61 among those without.13 Independent of diabe-
`
`Mayo Clin Proc. • December 2008;83(12):1373-1381 • www.mayoclinicproceedings.com
`
`1377
`
`MPI EXHIBIT 1142 PAGE 5
`
`

`

`EARLY DETECTION AND MANAGEMENT OF DIABETIC KIDNEY DISEASE
`
`No diabetes mellitus/no CKD (n=869,902)
`
`Diabetes mellitus/no CKD (n=179,777)
`
`No diabetes mellitus/CKD (n=25,573)
`
`Diabetes mellitus/CKD (n=17,949)
`
`CHF
`
`AMI
`
`CVA/TIA
`
`PVD
`
`ASVD
`
`Death
`
`60
`
`50
`
`40
`
`30
`
`20
`
`10
`
`0
`
`Incidence per 100 patient-years
`
`FIGURE. Rates of cardiovascular events in 2000-2001, per 100 patient-years, among individuals without diabetes
`mellitus or chronic kidney disease (CKD); with diabetes mellitus but without CKD; without diabetes mellitus but
`with CKD; and with both conditions. AMI = acute myocardial infarction; ASVD = atherosclerotic vascular disease;
`CHF = congestive heart failure; CVA/TIA = cerebrovascular accident/transient ischemic attack; PVD = peripheral
`vascular disease. Data from J Am Soc Nephrol.57
`
`tes status, microalbuminuria was shown to be an indepen-
`dent, continuous risk factor for CVD. Each increase in
`ACR of 0.4 mg/mmol increased the risk of a cardiovascular
`event by 5.9%.13
`In a study of more than 2500 men and women with no
`history of CVD, the relative risk of a cardiovascular event
`more than doubled when urinary albumin excretion was at
`least 15 µg/min.12 This increased risk was found to be
`independent of age, creatinine clearance rate, diabetes sta-
`tus, hypertension, total cholesterol level, and high-density
`lipoprotein cholesterol level.
`It is particularly striking that many patients with CKD,
`particularly elderly patients, may be several times more
`likely to die before progression to ESRD.60 Therefore, al-
`though 26 million patients in the United States have CKD,
`only a fraction will develop ESRD.
`Many clinicians assume that most patients with CKD
`die of CVD before progression to ESRD, but progression is
`more likely to happen in patients with DKD. Although an
`association exists between kidney disease and increased
`CVD risk, there are no controlled trials indicating whether
`treatment of patients with CKD improves CVD outcomes.
`Therefore, this association may be related to the common
`risk factors involved.
`
`TREATMENT STRATEGIES
`CVD Risk Reduction. On the basis of these and other
`findings, the scientific advisory boards of the American
`Heart Association and NKF Kidney Disease Outcomes
`Quality Initiative, as well as the Seventh Report of the Joint
`National Committee on Prevention, Detection, Evaluation,
`and Treatment of High Blood Pressure, recommend that
`
`patients with DKD be considered in the highest-risk group
`for CVD.56,61,62 Therefore, in addition to the slowing of
`kidney disease progression, treatment of patients with
`DKD should include efforts to manage cardiovascular risk
`factors such as hypertension, dyslipidemia, and hypergly-
`cemia on the basis of the more aggressive treatment goals
`recommended for patients at highest risk of CVD. Specifi-
`cally, in patients with DKD and CKD stages 1 to 4, recent
`clinical practice guidelines by the NKF Kidney Disease
`Outcomes Quality Initiative recommend a target blood
`pressure level of less than 130/80 mm Hg, a target low-
`density lipoprotein cholesterol level of less than 100 mg/
`dL, and a target HbA1c level of less than 7.0%.1
`Because most patients with DM and kidney disease also
`have hypertension and dyslipidemia, these treatment goals
`apply to most patients with DKD. However, it is important
`to remember that patients with advanced renal dysfunction
`were excluded from many of the studies on which these
`recommendations were based. Furthermore, targeting
`blood pressure, low-density lipoprotein cholesterol, and
`HbA1c at levels lower than suggested targets or applying
`such target goals to patients with stage 5 CKD (ie, ESRD)
`may not yield additional benefit in regard to CVD risk
`reduction and may actually place patients at increased risk
`of adverse events.63-65
`Slowing Progression of DKD. Activation of the renin-
`angiotensin-aldosterone system (RAAS) has long been rec-
`ognized as a key regulator of CKD progression. The RAAS
`is capable of causing kidney damage through several
`mechanisms, including systemic and glomerular hyperten-
`sion, increased glomerular capillary permeability, and local
`inflammation within the kidneys via release of several
`
`1378
`
`Mayo Clin Proc. • December 2008;83(12):1373-1381 • www.mayoclinicproceedings.com
`
`MPI EXHIBIT 1142 PAGE 6
`
`

`

`chemokines and profibrotic cytokines.66 Use of an angio-
`tensin-converting enzyme (ACE) inhibitor or an angio-
`tensin II receptor blocker (ARB) has been shown to retard
`GFR decline in patients with macroalbuminuria and either
`type 1 or type 2 DM.67-69 In addition, the mainstay of DKD
`management has long been inhibition of the RAAS with
`ACE inhibitors or ARBs.
`However, neither ACE inhibitors nor ARBs have been
`shown to reverse or even stabilize GFR in patient