H1AMR006 CKD 1_3/1-15-02

C.K.D.THE JOURNAL OF CHRONIC KIDNEY DISEASE

Editorial: Early referral of patients with kidney disease: overcoming the obstacles

Current care of patients with chronic kidney disease: Opportunities for improvement

One-on-One: An interview with Gerald Bernstein, MD, on the care of the diabetic nephropathy patient: a team effort toward optimal outcomes.

Lit Review:Selected papers from the literature

What’s new:Scarring may play major role in transplant rejection

Your Opinion: Treatment of hypertension

Current

care of

patients

with chronic

kidney

disease

Diagnostic Challenge:

See page 30 to review

this puzzling case.

Volume 1, Number 3

JANUARY 2002

7

11

15

22

24

28

30

Editorial: Early referral of patients with kidney disease:

overcoming the obstacles

By Steven Fishbane, MD, and Allen R. Nissenson, MD

Current care of patients with chronic kidney disease:Opportunities for improvement

By Allen R. Nissenson, MD, and Brian J.G. Pereira, MD, MBA

One-on-One: An interview with Gerald Bernstein, MD, on the

care of the diabetic nephropathy patient: a team

effort toward optimal outcomes.

Perspectives on diabetic nephropathyBy Eli A. Friedman, MD

Lit Review: Remission and regression in the nephropathy of type 1

diabetes when blood pressure is controlled aggressively

What’s New: Scarring may play major role in transplant rejection

Your Opinion: Treatment of hypertension

Diagnostic Challenge:

3

Table of ContentsTable of Contents

Submitted by Jill Lindberg, MD

Co-EditorsSteven Fishbane, MD

Mineola, New York

Allen R. Nissenson, MDLos Angeles, California

Editorial Advisory BoardAnatole Besarab, MDMorgantown, West Virginia

W. Kline Bolton, MDCharlottesville, Virginia

David N. Churchill, MDHamilton, Ontario, Canada

Robert N. Foley, MDSalford, United Kingdom

Raymond Hakim, MD, PhDNashville, Tennessee

Annamaria Kausz, MDBoston, Massachusetts

Alan S. Kliger, MDNew Haven, Connecticut

Adeera Levin, MDVancouver, British Columbia, Canada

Jill Lindberg, MDNew Orleans, Louisiana

Roger London, MDWhite Plains, New York

Brian J.G. Pereira, MD, DM, MBABoston, Massachusetts

Richard A. Sherman, MDNew Brunswick, New Jersey

John C. Stivelman, MDSeattle, Washington

Paul Zabetakis, MDOak Park, Illinois

Publishing StaffDavid Dunn

President

Keith J. CroesEditorial Director

Norma Padilla, PhDDirector, Scientific Affairs

Kerry MarianiSenior Project Director

Soleenne DesravinesProject Director, Scientific Affairs

Sandra SchallerArt Director

pproximately 6.2 million Americans have chronic kidney disease (CKD),1 andcare for these patients is often less than ideal.2 Medical management of CKDrequires attention to a variety of interrelated problems that have a progressiveimpact on patients’ health. For example, anemia and cardiovascular disease

(CVD), both common in CKD,2—4 are interrelated.2 Yet both are inadequately treated in manypatients.2, 5, 6 Anorexia and malnutrition are integral parts of the spectrum of conditions asso-ciated with CKD, yet patients rarely are referred for dietary consultation.7 In the latter stagesof CKD, planning for and placing vascular access are vitally important, yet most patientsstart dialysis without a functioning permanent vascular access.7 These examples illustrate the need for improvement in CKD care.

An important factor contributing to these less-than-ideal statistics may be delayed referralof patients by primary care physicians (PCPs) to nephrologists. In fact, many patients neversee a nephrologist until the 3 months leading up to dialysis initiation.7 This article exploresreasons for delayed referral, as well as potential solutions.

Advantages of nephrologist careWhy is early referral important? What data support the hypothesis that care by nephro-

logists offers advantages to patients with CKD?

Sesso and Belasco8 evaluated the effect of late diagnosis of CKD on mortality risk. Theyfound that late diagnosis led to a decrease in 6-month survival—from 87% to 69% (P<0.01).Similarly, studies indicate a 6-fold increase in hospital days with late referral,9 as well aslengthening of emergency treatment by 2- to 10-fold in inpatient stay over elective care.10

Other data indicate that failure to refer to a nephrologist adversely affects likelihood of vascu-lar access placement, need for emergency dialysis, hypertension and anemia management, and risk of pulmonary edema.9, 11

Taken together, a large number of studies are remarkably consistent in the finding ofimportant benefits when nephrologists are involved early in the care of patients with CKD.Although none of these studies are randomized clinical trials, the trend is compelling.

Reasons for delayed referralPCPs choose to refer patients to specialists based on their assessment of patients’ needs

and their level of confidence in their ability to adequately meet those needs. Until recently,the PCP had complete autonomy in referral decisions. The introduction of managed careplans as payers for primary care created obstacles to specialist referral.7,12 The administrativeprocesses set up by managed care companies undoubtedly have contributed to delayed referral to nephrologists.

In addition to managed care issues, a variety of other factors affect nephrology referral(Table 1). Some patients may have entered a late stage of disease by the time they see a PCP.For example, a young patient with immunoglobulin-A nephropathy may not be aware thathe has the disease until kidney function is severely diminished.9 Indeed, CKD is oftenasymptomatic even as it progresses, and perhaps is more likely to be detected earlier inolder individuals than in young, healthy people.

CKD—The Journal of Chronic Kidney Disease (ISSN 1532-4419) is published quarterly by HealthVizion Communications, Inc., and sponsored by Amgen Inc. Copyright 2002 HealthVizion Communications. All rights reserved.

Address correspondence and request for reprints to the Editorial Director, HealthVizion Communications, 20 Waterview Blvd., Parsippany, NJ 07054-1295; fax: 973-263-3952.

[ ]

Early referral of patients with kidney disease: overcoming the obstacles

Steven Fishbane, MDMineola, New York

Allen R. Nissenson, MDLos Angeles, CaliforniaEditorialEditorial

A

4

Table 1. Factors That Contribute to LateReferral to Nephrologists

Managed care limitations on specialty referral

Late presentation to the PCP

PCPs not recognizing that kidney disease is present because of:— failure to recognize CKD in older and smaller

patients with normal serum creatinine levels

— failure to appropriately screen high-risk patients,such as those with diabetes mellitus or hypertension

PCPs not fully understanding CKD treatment issuesand the importance of CKD-related complications

PCPs being unaware of the incremental benefit ofnephrology consultation

PCP’s fear of loss of control over their patient

PCP’s negative evaluation of previous CKD patientreferral to nephrologists

Inadequate communication by nephrologists

Patient reluctance to see another doctor

To get an idea of the relative weight PCPs assign to the various factors involved in referral decisions, we conducted a survey (unpublished) of 105 PCPs from Long Island, NY. Seventy-three provided detailed explanations of their perception of the role of nephrology consultation in CKD patients. Fifty-one of these PCPs reported that they occasionally referredpatients late in the course of CKD. Table 2 lists some of the reasons reported by the respondents.

Inadequate knowledge of issuesAmong the factors that can potentially delay referral

to the nephrologist is an inadequate knowledge ofCKD-related issues among some PCPs. The PCP mayfail even to recognize that CKD is present. Many PCPsrely solely on serum creatinine (SCr) levels as a screenfor kidney disease, despite the important diagnosticlimitations of the test.13, 14 A medium-size 70-year-oldwoman may have an SCr level of 1 mg/dL, a seeming-ly normal level, despite a creatinine clearance (Ccr) aslow as 40 mL/min, a rate that signals the presence ofsignificant kidney functional impairment. Thereappears to have been little educational effort to informPCPs about this pitfall of SCr levels. Moreover, the hec-tic pace of primary care practice makes it unlikely thatthe PCP will have time to calculate an estimated Ccr.

This problem could be solved if laboratories were to report both SCr and Ccr results. Laboratories couldutilize the Cockcroft-Gault formula, which has an 84%correlation with measured Ccr,13 and provide Ccr esti-mates in their standardized reports. Except for thepatient’s weight, the laboratory has access to all thedata necessary to do this. The Ccr estimate, therefore,could be reported as a range instead of a single figure.For example, in the case discussed above, the SCr levelof 1 mg/dL could be reported with a likely Ccr range of 33 to 66 mL/min for patients weighing 40 to 80 kg.Alternatively, a “normogram” could be displayed.

Knowledge of CKD-related issues extends also to thePCP’s ability to identify patients who need appropriatescreening. For example, PCPs may overlook micro-albumin testing in patients with diabetes mellitus, orSCr levels and urinalysis in patients with hypertension.A renal sonogram may be overlooked for the adultchild of a patient with polycystic kidney disease.

Even when PCPs recognize that CKD is present, theymay have incomplete knowledge of how it should bemonitored and treated. The importance of anemia,malnutrition, early bone disease, and vascular accessplanning may be underappreciated, leading to latereferral to the nephrologist.

Nephrologist responsePCPs often indicate that, after referring patients with

early CKD to nephrologists, patients sometimes reportthat nephrologists express bewilderment as to why apatient not needing dialysis is being referred to them.(Many of the PCPs in our survey described such experiences.) Why would nephrologists react this way?It seems clear that nephrologists could do much to con-tribute meaningfully to the care of such patients.

Two factors may help explain this response. First,some nephrologists may not be sufficiently aware ofthe important purpose of consultation in this setting,from the standpoints both of medical care and of building a successful practice. Second, some nephrolo-gists may simply be too busy to provide care to patients who are in the early stages of CKD. Indeed, a growingshortage of nephrologists may prompt PCPs, of necessity, to focus increasingly on the care of patientswith end-stage renal disease.15

Importance of communicationMany of the PCPs in our survey were disappointed

with the level of communication they typically

4

5

experience with nephrologists. Their complaints oftencited an inadequate explanation of tests that wereordered, and treatment and medication plans that wereprescribed. Consultation letters and phone calls to thePCP can effectively address this type of complaint.

Written communication by the nephrologist to thePCP should be structured in a way that adequatelyinforms the PCP of treatment plans and goals. The consultation note serves a broader purpose than simplyproviding medical information; the letter builds on thePCP’s perception of the nephrologist’s quality and com-pleteness of care. The letter also serves a marketingpurpose, enhancing the PCP’s sense of the importanceand value of nephrology consultation.

For patients with early CKD, the nephrologist shouldemphasize aspects of care that ought to be addressed,including blood pressure management, identificationand treatment of the CKD etiology, management ofanemia and early bone disease, and the interactionbetween CKD and existing comorbidity. Similarly, communication with the patient should reflect thevalue that the nephrologist’s involvement provides.

Many PCPs fear “losing” patients to nephrologists.Several issues likely stimulate this concern. The PCPmay have cared for the patient for many years, gainingthe patient’s trust and cultivating a healthy therapeuticrelationship. Specialist referral may be viewed as dif-fusing the strength of this relationship. The PCP mayfear the loss of future income if the patient decides toseek subsequent primary care from the nephrologist.Indeed, Campbell and colleagues16 found that 19% ofsurveyed PCPs expressed this concern. The PCP alsomay be worried that the nephrologist will make the

patient aware of treatments that should have, but werenot, previously provided.

Nephrologists, like any specialist, should take greatcare in their communication with referred patients notto diminish the patient’s perception of their PCP. Thenephrologist benefits by clearly expressing both to thePCP and the patient that the care they are providing isconsultative, and that the nephrologist is working incollaboration with the PCP.

Initiatives designed to inform PCPs of issues surrounding the care of CKD patients could go fartoward stimulating earlier referrals. In one interestingcollegial approach, a nephrology group invited PCPs intheir geographic area to an evening session. A nationalexpert lectured on principles of CKD care. Participantswere then seated with five PCPs and a nephrologist ateach table. Over dinner, the nephrologists led their PCPgroup through a structured discussion of CKD issues.The result was one of relationship building and anenhanced and shared understanding of the respectiveroles the nephrologist and PCP play in the care ofpatients with CKD.

SummaryEarlier referral of patients with CKD to nephrologists

is an important goal to achieve if patient treatment is to improve. A variety of dynamic issues are involved in this process. Certain trends, such as the worseningshortage of nephrologists, may hinder efforts toincrease the early referral of these patients. Strategiessuch as the use of nurse practitioners and physicianassistants, and disease-management approaches, offer promise. CKD

Table 2. Reasons Given by PCPs for Delaying Referral to Nephrologists*

Feel very capable of treating kidney-related problems 65% → 10%†

Have found little value in early nephrology consultation 80%

Fear of “losing” patient 34%

Financial referral issues 31%

Long delays to get appointments 25%

Old patients, comorbidity 20%

Fear of poor reflection on primary care 2%

*51 of 105 PCPs surveyed delay referral until creatinine clearance is <50 mL/min.†65% initially reported that they felt capable of treating these patients: after being asked to review their scores on the analog scale items, only 10% remained confident.

6

References1. Jones CA, McQuillan GM, Kusek JW, et al. Serum creatinine levels

in the US population: Third National Health and NutritionExamination Survey (erratum in Am J Kidney Dis. 2000;35:178). Am J Kidney Dis. 1998;32:992-999.

2. Obrador GT, Ruthazer R, Arora P, Kausz AT, Pereira BJ. Prevalence of and factors associated with suboptimal care before initiation ofdialysis in the United States. J Am Soc Nephrol. 1999;10:1793-1800.

3. Foley RN, Parfrey PS, Harnett JD, et al. Clinical and echocardio-graphic disease in patients starting end-stage renal disease therapy.Kidney Int. 1995;47:186-192.

4. Levin A, Singer J, Thompson CR, Ross H, Lewis M. Prevalent left ventricular hypertrophy in the predialysis population: identifying opportunities for intervention. Am J Kidney Dis.1996;27:347-354.

5. Kausz AT, Obrador GT, Pereira BJ. Anemia management in patients with chronic renal insufficiency. Am J Kidney Dis.2000;36:S39-S51.

6. McClellan WM, Knight DF, Karp H, Brown WW. Early detection and treatment of renal disease in hospitalized diabetic and hypertensive patients: important differences between practiceand published guidelines. Am J Kidney Dis. 1997;29:368-375.

7. Arora P, Obrador GT, Ruthazer R, et al. Prevalence, predictors, and consequences of late nephrology referral at a tertiary care center. J Am Soc Nephrol. 1999;10:1281-1286.

8. Sesso R, Belasco AG. Late diagnosis of chronic renal failure and mortality on maintenance dialysis. Nephrol Dial Transplant.1996;11:2417-2420.

9. Jungers P, Zingraff J, Page B, Albouze G, Hannedouche T, Man NK.Detrimental effects of late referral in patients with chronic renalfailure: a case-control study. Kidney Int. 1993;41:S170-S173.

10. Eadington DW. Delayed referral for dialysis. Nephrol DialTransplant. 1996;11:2124-2126.

11. Schmidt RJ, Domico JR, Sorkin MI, Hobbs G. Early referral and itsimpact on emergent first dialyses, health care costs, and outcome.Am J Kidney Dis. 1998;32:278-283.

12. Franks P, Williams GC, Zwanziger J, Mooney C, Sorbero M. Why dophysicians vary so widely in their referral rates? J Gen Intern Med.2000;15:163-168.

13. Cockcroft DW, Gault MH. Prediction of creatinine clearance fromserum creatinine. Nephron. 1976;16:31-41.

14. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A moreaccurate method to estimate glomerular filtration rate from serumcreatinine: a new prediction equation. Modification of Diet in RenalDisease Study Group. Ann Intern Med. 1999;130:461-470.

15. Bolton WK, Kliger AS. Chronic renal insufficiency: current under-standings and their implications. Am J Kidney Dis. 2000;36:S4-S12.

16. Campbell JD, Ewigman B, Hosokawa M, Van Stone JC. The timingof referral of patients with end stage renal disease. Dial Transplant.1989;18:660-686.

7

ore than 6 million Americans have elevat-ed serum creatinine (SCr) concentrations(≥ 1.5 mg/dL), and an estimated 2.5 mil-lion may have chronic kidney disease

(CKD) if CKD is defined as SCr levels of 1.7 mg/dL orgreater.1 Unfortunately, a number of studies haverevealed that patients with CKD are not routinely beingidentified in the early stages of their disease.2-5 In addi-tion, even after they have been diagnosed, many CKDpatients may not be receiving optimal treatment basedon current standards of care.2, 5, 6

This may prove to be penny wise and pound foolishin the long term. Early identification and comprehen-sive treatment of these patients may have a significantimpact not only on their quality of life, but also on theoverall cost of their care. The following steps are crucial:

Timely introduction of therapies to slow the progression of CKD

Identification and management of the comorbidi-ties commonly associated with CKD (eg, diabetes mellitus and heart disease)

Management of common complications of CKD (eg, anemia and renal osteodystrophy)

Smooth transitions to renal replacement therapywhen necessary.

We recently conducted a study of the patterns ofpractice for the treatment of 1,658 patients with CKD at a large, integrated health maintenance organization(HMO) in New Mexico.7 Because an extensive database,with considerable patient-specific information, wasavailable for analysis, we were able to gain insights intothe health care resources used by patients with CKD,including their patterns of hospitalization, use of vari-ous types of outpatient providers, and use of outpatientprescription medications. These observations haveadded new information to our understanding of howpatients with CKD are being treated.

Erythropoietin useAlthough the prevalence of anemia was high among

the patients with CKD in our cohort, recombinant

human erythropoietin (rHuEPO) was used infrequent-ly in 7.4% of patients. The majority of patients whoreceived rHuEPO exhibited moderate or severe anemia, and very few patients with hematocrit (Hct)values of >32.9% received this medication.7 For themost part, the patients who received rHuEPO alsoexhibited the most severe CKD on the basis of SCr val-ues. In addition, at any given Hct level, the likelihoodthat a patient would be treated with rHuEPO increasedwith increasing SCr values. Finally, it was uncommonfor patients to receive rHuEPO unless patients hadbeen seen by a nephrologist, even after controlling forCKD severity and Hct level. More detailed analyses of anemia and its treatment in this cohort are ongoing.

ACE inhibitor useAs shown in a recent meta-analysis,8 angiotensin-

converting enzyme (ACE) inhibitor use among patientswith CKD is effective in slowing the progression ofrenal disease, whether or not the patient also suffersfrom diabetes mellitus (DM). A recent analysis byGolan and colleagues9 demonstrated that the use ofACE inhibitor therapy for patients with type 2 DM isalso cost-effective. There were several noteworthy findings regarding the use of ACE inhibitors in ourcohort7 (see Table 1):

Use of ACE inhibitors among patients with CKD was disappointingly low, particularly amongpatients with SCr values of ≥ 4 mg/dL. During thefirst year after entry into the study, fewer thanhalf of the patients with CKD had been prescribedan ACE inhibitor.

When SCr values were ≤ 3.9 mg/dL, the percent-age of patients who received an ACE inhibitorwas similar regardless of SCr level, but this percentage decreased to less than 20% whenSCr levels were ≥ 4 mg/dL.

Although the likelihood of ACE inhibitor use washigher among diabetic patients with CKD thanamong nondiabetic patients with CKD, ACEinhibitor use among diabetic patients was stilllow, regardless of their SCr values, ranging from

Current care of patients with chronic kidney disease:Opportunities for improvementBetter adherence to practices known to be of clinical benefit not only will improve patient outcomes, but also may reduce the cost of care.

featurefeatureAllen R. Nissenson, MD

Los Angeles, California

Brian J.G. Pereira, MD, MBABoston, Massachusetts

Author information: Author information: Allen R. Nissenson, MD, is Professor of Medicine at the University of California, Los Angeles, and Director of the UCLA Dialysis Program.Brian J.G. Pereira, MD, MBA, is Senior Vice President of the New England Medical Center, and Professor of Medicine at Tufts University School of Medicine and the Sackler School of Graduate Biomedical Sciences. Dr. Nissenson is supported in part by the Richard Rosenthal Dialysis Fund.

M

8

45% to 59% for SCr values below 4.0 mg/dL,down to 28% for SCr values of 4.0 mg/dL andabove.

Perhaps even more disappointing was the observation that the likelihood of ACE inhibitoruse was no higher among patients with CKD who were cared for by a nephrologist than by a primary care physician (PCP), even after controlling for CKD severity.

Although previous reports10—13 documented theeffects of late referral of patients with CKD to nephrol-ogists, this study provides a more complete view of thecurrent practice patterns for these patients, including a number of aspects of care not previously well characterized.

Comorbid conditionsPatients with CKD often suffer from other serious

conditions as well. Careful attention to the manage-

ment of these associated comorbid conditions canreduce hospitalizations for these patients. Cardio-vascular disease is highly prevalent among patientswith CKD and often leads to hospitalization.14, 15 Closerattention to the management of heart disease in thispopulation, as outlined by Meyer and Levey in aNational Kidney Foundation report,16, 17 could substan-tially improve outcomes. In addition, several studiesdemonstrated that even partial correction of anemiawith rHuEPO (to a hemoglobin level of 11 to 12 g/dL)among dialysis patients may prevent left ventricularhypertrophy from developing or may lead to its regression.18, 19

In 1994, a National Institutes of Health ConsensusDevelopment Conference on CKD20 recommended thata nephrologist be consulted early in the course of renaldisease (when SCr values reach 1.5 mg/dL for womenor 2 mg/dL for men) so that appropriate diagnostictests can be performed and interventions, when avail-able, can be prescribed.

Table 1. Percentage of Patients Who Received an ACE Inhibitor, According to SCr Category and Whether They Were Seen by a Nephrologist

Annualized Percentage

SCr Category All Patients Patients Seen by Nephrologist* Patients Not Seen byNephrologist

All patients n=1,658 n=317 n=1,341

All (n=1,658) 0.34 0.31 0.35

<2.0 mg/dL (n=1,203) 0.36 0.38 0.36

2.0 to 2.9 mg/dL (n=289) 0.31 0.31 0.31

3.0 to 3.9 mg/dL (n=53) 0.43 0.42 0.45

≥4.0 mg/dL (n=113) 0.15 0.16 0.09

Diabetic patients n=425 n=109 n=316

All (n=425) 0.49 0.44 0.51

<2.0 mg/dL (n=287) 0.50 0.56 0.50

2.0 to 2.9 mg/dL (n=75) 0.59 0.57 0.60

3.0 to 3.9 mg/dL (n=20) 0.45 0.43 0.56

≥4.0 mg/dL (n=43) 0.28 0.28 0.29

Nondiabetic patients n=1,233 n=208 n=1,025

All (n=1,233) 0.29 0.24 0.30

<2.0 mg/dL (n=916) 0.32 0.32 0.32

2.0 to 2.9 mg/dL (n=214) 0.21 0.19 0.22

3.0 to 3.9 mg/dL (n=33) 0.41 0.41 0.41

≥4.0 mg/dL (n=70) 0.05 0.05 0.06*Seen by a nephrologist within 1 year after entry into the CKD cohort.

9

In our study,7 however, only a minority of patientswith CKD were seen by a nephrologist during the firstyear of observation after entry into the study (Table 2).Although the percentage of patients who visited anephrologist during the first year of the studyincreased with increasing SCr values, nephrologiststended not to be involved in the care of patients whoseSCr values were lower than 3 mg/dL.

Unfortunately, the experience we observed is typicalof clinical practice throughout the United States.Among another cohort of patients who were over theage of 65 before the initiation of dialysis, 81% had notbeen seen by a nephrologist until 3 months beforebeginning dialysis, and 47% had not been seen by anephrologist until 1 month before dialysis.2 Two otherfindings from our study7 are worth noting: Diabetesmellitus, hypertension, or both were significantly associated with increased SCr and were increasinglyprevalent with higher SCr values. In addition, the number of hospitalizations increased directly with thelevel of SCr (Table 3), and the majority of hospital-izations were related to associated comorbidities, rather than to CKD itself.

DiscussionThe consequences of these late referrals to a

nephrologist are devastating. Patients who are referred

late are more likely to present for dialysis with pul-monary edema, the need for temporary vascularaccess, hypoalbuminemia, severe uremia,10 and meta-bolic acidosis.12 Late referrals of patients to nephrologistfurther increase the need for invasive therapeuticmeasures, clinical morbidity, and, ultimately, costlyhospitalization.12 Several studies have demonstratedthat late referral increases morbidity and costs forpatients with CKD.10-13

Even when patients are referred to a nephrologist,some aspects of care that have proven beneficial forpatients with CKD are not adhered to consistently. For example, there was no evidence in our study thatpatients with CKD who visited a nephrologist had ahigher likelihood of ACE inhibitor use than did thosewho did not visit a nephrologist.7 In addition, signifi-cant numbers of anemic patients with CKD who visited a nephrologist did not receive rHuEPO. Ourfindings are consistent with practices observed in other settings.10, 21

Although in some cases this may reflect inadequateresources to pay for this costly medication, this is notthe explanation for many patients. Rather, the physi-cian, whether a PCP or a nephrologist, has simply nottreated the anemia. Similarly, McClellan and col-leagues21 demonstrated that ACE inhibitors are infre-

Table 3. Hospital Utilization by Patients With CKD

SCr Category Hospitalization Utilization(Average No./Patient-Yr)

All (n=1,658) 0.84

<2.0 mg/dL (n=1,203) 0.72

2.0 to 2.9 mg/dL (n=289) 0.96

3.0 to 3.9 mg/dL (n=53) 1.20

≥4.0 mg/dL (n=113) 1.56

Table 2. Percentage of Patients Who Visited a PCP or Nephrologist, According to SCr Category

SCr Category Annualized PercentagePCP* Nephrologist

All (n=1,658) 0.88 0.22

<2.0 mg/dL (n=1,203) 0.93 0.12

2.0 to 2.9 mg/dL (n=289) 0.86 0.31

3.0 to 3.9 mg/dL (n=53) 0.82 0.72

≥4.0 mg/dL (n=113) 0.50 0.80

PCP = primary care physician*The type of provider was assessed on the basis of specialty-specific location-of-service codes used by the health maintenance organization.

quently prescribed for patients with CKD who arebeing discharged from the hospital. The appropriateuse of rHuEPO and ACE inhibitors could substantiallyimprove outcomes for this patient population.

For monitoring of nutritional status, effects ofreduced-protein diets, use of appropriate multivita-mins, and control of phosphorus and sodium intake,the routine availability of a trained dietitian is essentialfor the optimal treatment of patients with CKD.

The results of our study suggest numerous opportu-nities to improve outcomes for this patient population.Better adherence to practices known to be of clinicalbenefit not only will improve patient outcomes but alsomay reduce the cost of care. Because of the high clini-cal and financial costs of ESRD programs, providersand policy-makers should view CKD as a major publichealth problem and should initiate innovative pro-grams to address this growing patient population. CKD

References1. Jones CA, McQuillan GM, Kusek JW, et al. Serum creatinine levels

in the US population: Third National Health and NutritionExamination Survey. Am J Kidney Dis. 1998;32:992-999 (erratum in Am J Kidney Dis. 2000;35:178)

2. Obrador GT, Pereira BJ. Early referral to the nephrologist and timelyinitiation of renal replacement therapy:a paradigm shift in the man-agement of patients with chronic renal failure. Am J Kidney Dis.1998;31:398-417.

3. Eadington DW. Delayed referral for dialysis. Nephrol DialTransplant. 1996;11:2124-2126.

4. Levin A. Anaemia in the patient with renal insufficiency: docu-menting the impact and reviewing treatment strategies. NephrolDial Transplant. 1999;14:292-295.

5. Jungers P. Screening for renal insufficiency: Is it worth while? Is itfeasible? Nephrol Dial Transplant. 1999;14:2082-2084.

6. Obrador GT, Ruthazer R, Arora P, Kausz AT, Pereira BJ. Prevalenceof and factors associated with suboptimal care before initiation ofdialysis in the United States. J Am Soc Nephrol. 1999;10:1793-1800.

7. Nissenson AR, Collins AJ, Hurley J, Petersen H, Pereira BJ,Steinberg EP. Opportunities for improving the care of patients withchronic renal insufficiency: current practice patterns. J Am SocNephrol. 2001;12:1713-1720.

8. Kshirsagar AV, Joy MS, Hogan SL, Falk RJ, Colindres RE. Effect ofACE inhibitors in diabetic and nondiabetic chronic renal disease: asystematic overview of randomized placebo-controlled trials. Am JKidney Dis. 2000;35:695-707.

9. Golan L, Birkmeyer JD, Welch HG. The cost-effectiveness of treat-ing all patients with type 2 diabetes with angiotensin-convertingenzyme inhibitors. Ann Intern Med. 1999;131:660-667.

10. Arora P, Obrador GT, Ruthazer R, et al. Prevalence, predictors, andconsequences of late nephrology referral at a tertiary care center. J Am Soc Nephrol. 1999;10:1281-1286.

11. Sesso R, Belasco AG. Late diagnosis of chronic renal failure andmortality on maintenance dialysis. Nephrol Dial Transplant.1996;11:2417-2420.

12. Jungers P, Zingraff J, Page B, Albouze G, Hannedouche T, Man NK.Detrimental effects of late referral in patients with chronic renalfailure: a case-control study. Kidney Int. 1993;41:S170-S173.

13. Schmidt RJ, Domico JR, Sorkin MI, Hobbs G. Early referral and itsimpact on emergent first dialyses, health care costs, and outcome.Am J Kidney Dis. 1998;32:278-283.

14. Xia H, Ebben J, Ma JZ, Collins AJ. Hematocrit levels and hospital-ization risks in hemodialysis patients. J Am Soc Nephrol. 1999;10:1309-1316.

15. Collins AJ, Li S, St Peter W, et al. Death, hospitalization, and eco-nomic associations among incident hemodialysis patients withhematocrit values of 36 to 39%. J Am Soc Nephrol. 2001;12:2465-2473.

16. Meyer KB, Levey AS. Controlling the epidemic of cardiovasculardisease in chronic renal disease: report from the National KidneyFoundation Task Force on cardiovascular disease. J Am SocNephrol. 1998;9(suppl 12):S31-S42.

17. Levey AS, Beto JA, Coronado BE, et al. Controlling the epidemic ofcardiovascular disease in chronic renal disease: What do we know?What do we need to learn? Where do we go from here? NationalKidney Foundation Task Force on Cardiovascular Disease. Am JKidney Dis. 1998;32:853-906.

18. Macdougall IC, Lewis NP, Saunders MJ, et al. Long-term cardiores-piratory effects of amelioration of renal anaemia by erythropoietin.Lancet. 1990;335:489-493.

19. Martinez-Vea A, Bardaji A, Garcia C, Ridao C, Richart C, Oliver JA.Long-term myocardial effects of correction of anemia with recom-binant human erythropoietin in aged patients on hemodialysis.Am J Kidney Dis. 1992;19:353-357.

20. National Institutes of Health Consensus Development ConferencePanel N. Morbidity and mortality of renal dialysis: an NIH consensus conference statement. Consensus DevelopmentConference Panel. Ann Intern Med. 1994;121:62-70.

21. McClellan WM, Knight DF, Karp H, Brown WW. Early detectionand treatment of renal disease in hospitalized diabetic and hyper-tensive patients: important differences between practice and pub-lished guidelines. Am J Kidney Dis. 1997;29:368-375.

10

11

Of US patients who reach end-stage renal disease (ESRD),more than 40% do so because of diabetic nephropathy.1 Asthe diabetes in most of these patients is already being treat-ed, endocrinologists are in a position to diagnose chronic kidney disease (CKD) in its early stages, before it has progressed to ESRD.

Blood pressure control, optimization of angiotensin-convert-ing enzyme inhibitor therapy, anemia treatment, treatment ofbone disease, attention to nutrition, selection of renal replace-ment modality in late-stage disease, and vascular access planning all are important aspects of care in all CKDpatients. Data from the US Renal Data System (USRDS),however, reveal notable deficiencies in such care. In a recentstudy2,3 utilizing USRDS data, Obrador and colleagues foundthat 60% of patients had albumin concentrations below thelower limit of normal when they began dialysis. According to another study,1,4 only 44% of patients had a functioningpermanent vascular access before starting dialysis, and 27%were still using a temporary access 30 days after dialysis initi-ation. These figures indicate that patients have not had well-coordinated care in the time leading up to dialysis. This is trueof both diabetic and nondiabetic patients. The opportunity toimprove care is particularly feasible in diabetics, however,since they are often being followed by primary care physicians(PCPs) or endocrinologists when kidney disease first begins.

CKD induces first mild and then severe anemia, regardlessof the primary cause of kidney functional loss. In diabeticnephropathy, the onset of anemia caused by erythropoietindeficiency can occur early, well before ESRD.5 Recent experience with the “glitazones” (thiazolidinediones) has raisedthe awareness of anemia among endocrinologists; still, onlyabout 20% of patients who begin dialysis with a hematocrit(Hct) level of 30% or less have received recombinant humanerythropoietin (rHuEPO) before reaching ESRD.3

CKD—THE JOURNAL OF CHRONIC KIDNEY DISEASEinterviewed Gerald Bernstein, MD, on issues surrounding the care and referral of patients with diabetic nephropathy. A former president of the American Diabetes Association,Bernstein is currently an attending physician at Beth IsraelMedical Center in New York.

CKD: According to US Renal Data System(USRDS) data,1 less than half ofpatients with end-stage renal disease

(ESRD) had a permanent vascularaccess before starting dialysis, andalmost one third were still using temporary access 30 days after dialysis initiation. Too many of these patientsalso seemed to have albumin, serumcreatinine (SCr), and hemoglobin (Hb)levels that, in an ideal world, wouldhave been better controlled. Whydon’t people get better care prior to dialysis?

GERALD BERNSTEIN, MD: I think it goes backto what we were taught—that once the process of kidney failure starts, it's inexorable and will progress.Although chronic kidney disease (CKD) cannot be prevented from progressing, you can slow the process dramatically with such things as angiotensin-convert-ing enzyme (ACE) inhibitors, a proper diet, blood pressure management, management of diabetes, etc.These aspects of care for patients with CKD are probably not emphasized enough in general medicine.

CKD: Do comfort levels have anything to dowith it? It’s rational to speculate, forinstance, that PCPs are comfortabletreating hypertension and prescribingACE inhibitors. It’s less likely theywould be comfortable treating CKD-related anemia or bone and mineralmetabolic disorders, or providingnutrition and dialysis education.

BERNSTEIN: Even if they have some degree of comfort, they may not be treating aggressively enoughcompared with someone who provides those servicesall the time in order to reach certain targets.

CKD: According to a study by Obrador andcolleagues,3 80% of patients initiatingdialysis with hematocrit (Hct) levelsless than 28% had not received treat-ment with rHuEPO. Does that findingsurprise you?

One-on-OneOne-on-One

Care of the diabetic nephropathy patient: a team effort toward optimal outcomesA large percentage of people with chronic kidney disease is diabetic. Their optimal treatment will require overcoming current gaps in medical education.

Gerald Bernstein, MDNew York, New York

12

BERNSTEIN: No, it doesn’t surprise me. As a familypractitioner or internist, you learn certain things aboutthe natural history of kidney disease. The understand-ing is that it is natural for someone whose SCr levelbegins to rise to have a fall in Hct level. And when that happens, the individual compensates through biochemical mechanisms to deliver more oxygen to the tissues. For most physicians, that’s where it leavesoff. It certainly did with me.

The typical approach for most physicians—if thepatient’s guaiac test is negative and bleeding can beruled out—is simply to observe the patient. If the SCrlevel is 2.5 mg/dL, that’s no big issue because thepatient is not ready for dialysis. Why do I need to make a referral if the patient has no specific signs orsymptoms related to SCr level?

CKD: Somewhere along the line it appearsyou developed a different view of thisscenario. When did that happen?

BERNSTEIN: A number of years ago. Having a largediabetes practice, I’ve been very much involved withnephrologists. We’ve had many patients who have pro-gressed to dialysis over 20 or 30 years, so I was nostranger to dialysis and the issues of vascular access.

For us the question was, when do we refer? For themost part, we didn’t refer to nephrologists until SCrwas up around 5 mg/dL. If the Hb level dropped to 8 g/dL, well, we were not alarmed because that’s whatwe were taught happens to these people. If they’refunctioning and going to work and don’t feel badly, and they’re not unusually acidotic or have a variety ofother clinical issues, then there’s no need to make areferral. When their SCr level got to around 5 mg/dL,we figured that even in the best of situations it mightbe another year or so before they would need dialysis.We then would make the referral for the access—andthis was usually before even the nephrologist wasready to address the access.

What began to become apparent as I watchednephrologists was that nephrologists were much morefocused on achieving certain goals, and the goals werebeyond what I could have achieved. My office wasn’tset up to accomplish these goals, nor was I attuned tothe needs of the patient in the same preventative wayduring the early stages kidney of disease evident by rising SCr levels.

A number of years ago, I began to change my prac-tices and started making referrals when SCr levels werebetween 2.5 and 3 mg/dL. I never made a referral

because of a drop in Hb level, nor do I know manyphysicians who have.

CKD: Did you ever take any steps to treat anemia?

BERNSTEIN: No, because it was regarded as naturalpathophysiology.

CKD: In the past several years, in your opinion, has the awareness of anemiaincreased, decreased, or remained the same?

BERNSTEIN: For the most part, I think it hasremained the same. The sophisticated medical commu-nity is giving it more attention, but I don’t think that’sthe case in the general medical community.

CKD: According to a consensus statementfrom the National Institutes of Health,6

referral to a nephrologist for CKD isrecommended at SCr levels of 1.5 and 2.0 mg/dL for women and men,respectively. Do you agree that peoplewith this level of kidney functionalimpairment should be referred to a nephrologist?

BERNSTEIN: Most nephrologists probably would notincorporate those people into a major part of their pro-gram. As I mentioned, I referred patients relativelyearly—when SCr levels were no higher than 3 mg/dL.I’ve worked with some well-respected nephrologists,and none has ever suggested that I should be referringearlier. And these were nephrologists with whom ourcommunications were very strong; it’s not as thoughthey were afraid of insulting me.

CKD: Based on the National KidneyFoundation Kidney Disease OutcomesQuality Initiative (NKF K/DOQI) guide-lines, however, they probably shouldbe urging earlier referral. Perhapsmore educational efforts are needed in nephrology?

BERNSTEIN: I think the problem has to be docu-mented and sold more aggressively. There are manymedical conditions that were not treated in the pastbecause the treatments were not available, or at leastnot generally available. The more medical careadvances, the more tools that you have and the moreproblems become treatable at an earlier stage.

Gerald Bernstein, MD

is a former president

of the American

Diabetes Association.

He is currently an

attending physician at

Beth Israel Medical

Center in New York.][

13

That’s where we are now with anemia; it is becom-ing more widely recognized that earlier treatment ofanemia may make a clinical difference in a patient’slife. In diabetes, we’re always dealing with quality-of-life (QOL) issues and the aggressive need to controlblood sugar. There are good data7 showing that bloodsugar control is associated with significant QOLimprovements. But a lot of people don’t know thattheir QOL is not so great because nobody asks them,and they assume everybody feels the way they feel.

CKD: The use of Hct and SCr measurementsis in question. The recent NKF-K/DOQIguidelines8 suggest that Hb should beused instead of Hct as a quantifier ofanemia in CKD patients. Additionally,it’s recognized that SCr levels underes-timate kidney functional impairment inmany patients9; consequently, thereappears to be increasing support forthe routine use of formulas that esti-mate glomerular filtration rate (GFR).10, 11

Are these messages being conveyedto the general practitioner andendocrinologist?

BERNSTEIN: Not really. I read at least three generalmedical journals, and these issues are almost never discussed.

Most of us some time ago learned that there was astraight-line relationship between SCr and GFR. Again,in the more sophisticated metabolic community, we’vebegun to understand that such a straight-line relation-ship does not always hold true. But in the general med-ical community, no, this is not widely appreciated.Most physicians don’t have the opportunity to interactwith “superspecialists” to get this kind of information. I would suspect if you looked at a series of grandrounds held over the course of the last year or two,there would be little or no discussion of this.

CKD: There’s also growing evidence ofadverse effects associated with Hb levels less than 10 g/dL in people withCKD even prior to ESRD.3, 8 Is it fair tosay that this information also is notwidely appreciated?

BERNSTEIN: Yes, that’s fair to say.

CKD: Many of these adverse effects are subclinical. Is that why anemia is notbeing picked up earlier?

BERNSTEIN: I’m sure that’s a factor. Subclinicaladverse effects are also a major part of diabetes care.Even a slight elevation in blood glucose will act as apoison over a period of 10 or 20 years and lead to kidney disease. Yet patients may not have any symp-toms—or think they don’t have any symptoms. Then it depends on the degree of pursuit.

One area that is a bit similar is diabetic stomach disease. These people often are symptomatic, but theirsymptoms fall within a range of minor complaints thatseem normal. Patients say, “Well, I've got an upsetstomach, I’ve got a little reflux, I feel bloated,” that kind of thing. They might talk to someone who doesn’thave diabetes who has the same complaints. It doesn’tappear to the patient to be an issue until it reaches an advanced stage.

CKD: What do we know about diabetes control and its effect on CKD?

BERNSTEIN: We know that controlling both bloodglucose and blood pressure can slow the progression ofCKD.12 There’s some interesting unpublished data indi-cating that people with even advanced kidney diseasemay, if they live long enough, demonstrate reversal of some of their kidney disease if their diabetes isextremely well controlled in that interval.

A recent study by Bosman and colleagues5 showedthat people with diabetes develop anemia earlier thanpeople without diabetes, given the same SCr levels.That’s an important observation that I’d like to see corroborated.

CKD: You seem to have developed a goodrelationship with nephrologists. Butyou also have been in a position toknow what others have experienced in caring for their patients with dia-betic nephropathy. Are there somegeneral lessons that you’ve learned in terms of referral to and interactionwith nephrologists?

BERNSTEIN: It’s important to remember that thenephrologist has a team. The nephrologist works with anutritionist the same way the endocrinologist workswith a nutritionist. The practicing physician needs todelegate more of the responsibility for certain elementsof the care of patients to nonphysicians—the diabeteseducator and the kidney educator—who then canspend the appropriate amount of time sorting out whatneeds to be done and getting the patient on the right

14

path. The physician simply doesn’t have time to do this in the ordinary working day.

The earlier the intervention, the better. We needaggressive nephrologists who will bring in patientseven if the patients don’t need to be seen often, andwho will get patients involved with the educationprocess. The nephrology educator can then monitorthem for such developments as diminished Hb level or a slow increase in SCr level, and alert the nephr-ologist as to when patients need additional medicaltreatment. CKD

References1. US Renal Data System. USRDS 1999 annual data report. Bethesda,

Md: National Institute of Diabetes and Digestive and KidneyDiseases, National Institutes of Health; 1999.

2. US Renal Data System. USRDS 1996 annual data report. Bethesda,Md: National Institutes of Health, National Institutes of Diabetes andDigestive and Kidney Disease; 1996.

3. Obrador GT, Ruthazer R, Arora P, Kausz AT, Pereira BJ. Prevalence ofand factors associated with suboptimal care before initiation of dialy-sis in the United States. J Am Soc Nephrol. 1999;10:1793-1800.

4. National Institute of Diabetes and Digestive and Kidney Diseases.Healthy People 2010: Chronic Kidney Disease. 2000.

5. Bosman DR, Winkler AS, Marsden JT, et al. Anemia with erythropoi-etin deficiency occurs early in diabetic nephropathy. Diabetes Care.2001;24:495-499.

6. National Institutes of Health. Morbidity and mortality of dialysis: anNIH consensus statement. NIH Consensus Statement. United States:National Institutes of Health; 1993:1-33.

7. Testa MA, Simonson DC. Health economic benefits and quality oflife during improved glycemic control in patients with type 2 dia-betes mellitus: a randomized, controlled, double-blind trial. JAMA.1998;280:1490-1496.

8. National Kidney Foundation. NKF-K/DOQI Clinical PracticeGuidelines for Anemia of Chronic Kidney Disease: update 2000. Am J Kidney Dis. 2001;37(suppl 1):S182-S238.

9. Jones CA, McQuillan GM, Kusek JW, et al. Serum creatinine levelsin the US population: Third National Health and NutritionExamination Survey. (erratum in Am J Kidney Dis. 2000;35:178)Am J Kidney Dis. 1998;32:992-999.

10. Cockcroft DW, Gault MH. Prediction of creatinine clearance fromserum creatinine. Nephron. 1976;16:31-41.

11. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A moreaccurate method to estimate glomerular filtration rate from serumcreatinine: a new prediction equation. Modification of Diet in RenalDisease Study Group. Ann Intern Med. 1999;130:461-470.

12. Zabetakis PM, Nissenson AR. Complications of chronic renal insufficiency: beyond cardiovascular disease. Am J Kidney Dis.2000;36:S31-S38.

15

Diabetic nephropathy is the leading cause of end-stagerenal disease (ESRD). Clinicians and researchers are opti-mistic, however, that new therapies may make diabetesmore manageable and limit its impact on the kidneys.

ccording to the World HealthOrganization, type 2 diabetes mellitusmay afflict 300 million individuals world-wide by 2025.1 In the United States,

according to the National Institutes of Health’s NationalDiabetes Information Clearinghouse, 15.7 million peo-ple—almost 6% of the population—have diabetes.2 Onethird of these people are unaware of their disorder.Diabetes was the seventh leading cause of death listedon US death certificates in 1996, contributing to193,140 deaths. More than 798,000 new cases are diag-nosed each year. Health care expenditures for diabetes-related problems amount to a minimum of $98 billionannually in the United States, and may be even higher.2

Diabetic nephropathy is the leading cause of end-stage renal disease (ESRD) in the United States, Japan,and most of industrialized Europe.3 As noted in the USRenal Data System (USRDS) 2001 annual report, morethan 40% of all ESRD patients had diabetes at initiationof therapy.4 Although much attention has been given tokidney disease in type 1 diabetes, the prevalence ofkidney disease caused by type 2 diabetes is significant.3

Contrary to widely prevalent thinking, kidney diseaseis as likely to develop in type 2 diabetes as in type 1.5

While there are differences in the prevalence of diabeticnephropathy among racial and ethnic groups,3 otheraspects of the two disorders—particularly the progres-sion of nephropathy—are remarkably similar.6

Diabetic patients with kidney disease have the worstoverall age-adjusted survival rate of all patients withESRD.4 When compared to patients with other causesof ESRD, diabetic patients have greater mortality andmorbidity due to comorbid disorders, especially cardio-vascular and cerebrovascular disease.7 Given thesefacts, the prognosis for patients with kidney diseaseand diabetes seems uniformly grim. However, there iscause for considerable optimism. Continuing advancesin understanding the pathogenesis of diabetic kidneydisease, and the development of new therapies based

on this understanding, may soon make the task of car-ing for diabetic patients with major complications farless daunting. The hope is that, within the very nearfuture, the prognosis for diabetic vasculopathy andnephropathy will be transformed from hopeless, inex-orable deterioration to a manageable metabolic disor-der permitting full function and even cure.

This article discusses the options available to the dia-betic ESRD patient and the need for overall diabeticcare during uremia therapy.

Options for ESRD treatment in diabetesDiabetic ESRD patients are managed similarly to

nondiabetic ESRD patients with two exceptions: First,simultaneous pancreas and kidney transplantation is a diabetes-specific therapy. Second, in our experi-ence, diabetic patients are more likely than nondiabeticpatients to decline treatment, an option we refer to aspassive suicide.

ESRD in diabetic persons reflects the demographicsof diabetes per se in that the incidence is higher inAfrican Americans, Hispanics, and Native Americans,3,

4, 8 and the peak incidence of ESRD occurs from the 5thto the 7th decade.8, 9 Blacks older than age 65 years facea 7-fold greater risk of diabetes-related kidney failurethan do whites.8

Type 2 diabetes accounts for the majority of patientswho require renal replacement therapy (RRT)4; how-ever, there is often confusion over diabetes type.10

Consequently, literature reports of the outcome ofESRD therapy by diabetes type are few and imprecise.Vasculopathic complications of diabetes, such as hyper-tension, have approximately the same impact on theprogression of diabetic nephropathy in type 2 diabetesas in type 1 diabetes.6

Depending on age, severity of comorbid disorders,available local resources, and patient preference, theuremic diabetic patient may be managed according to different protocols (Table 1). Unfortunately, in theso-called preterminal care of diabetic patients withESRD, inadequate attention often is paid to the control of hypertension, hyperlipidemia, anemia, and ophthalmologic problems.

Perspectives on diabetic nephropathyAuthor information: Eli A. Friedman, MD, is Distinguished Teaching Professor of Medicine and Chief of the Renal Disease Division, Department of Medicine, State University

of New York Health Science Center at Brooklyn, NY. He is a recipient of both the American Association of Kidney Patients Medal of Excellence and theAmerican Kidney Fund’s National Torchbearer Award.

featurefeature By Eli A. Friedman, MDBrooklyn, New York

A

16

Table 1. Options in Uremia Therapy for Diabetic ESRD Patients

1. No specific uremia intervention (passive suicide)

2. Peritoneal dialysis

Intermittent peritoneal dialysis

Continuous ambulatory peritoneal dialysis

Continuous cyclic peritoneal dialysis

3. Hemodialysis

Facility hemodialysis

Home hemodialysis

4. Kidney transplantation

Cadaver donor kidney

Living donor kidney

5. Pancreas plus kidney transplantation(From Friedman,199911).

For the large majority—more than 80% of diabetic persons who develop ESRD in the United States—main-tenance hemodialysis (HD) is the only renal replace-

ment regimen that will be employed.11 Approximately12% of diabetic persons with ESRD will be treated byperitoneal dialysis (PD), while the remaining 8% willreceive a kidney transplant.11

Rehabilitation is difficult in diabetic ESRD patients. We have found that a couple of factors are especiallyimportant. First, patients tend to fare best when theyparticipate in the design of their own treatment regi-men. Second, a functioning kidney transplant permitsmarkedly superior rehabilitation than that attained by either HD or PD.

HemodialysisWe have found that as many as 55% of diabetic HD

patients were unable to perform routine living choreswithout assistance vs 25% of nondiabetic HD patients.12

In one study of 430 patients, the percentage of individ-uals requiring a wheelchair was almost 3-fold higher indiabetic vs nondiabetic patients: 29% vs 10%.12 Otherstudies have confirmed these observations.11 Theinescapable conclusion of studies to date is that main-tenance HD does not permit a full return to life'sresponsibilities for most diabetic individuals.

Survival of Diabetic ESRD Patients

USRDS 2000 Years After Initiating Treatment

Percent Surviving Live DonorTransplant

Cadaver DonorTransplant

10

95.4%

93.1%

93.1%

78.4% 75.4%

61.6%

23.1% 3.8%

39.4%

42.5%

71.0%

2 5 10

Dialysis

100

80

60

40

20

0

86.5%

Figure 1: Long-term survival of diabetic ESRD patients shows the grim chances of surviving for a decade on dialysis (data for HDand PD are pooled). Fewer than 1 in 25 diabetic patients treated by dialysis will live for 10 years.9

17

Table 2. Comparison of ESRD Options for Diabetic Patients

Factor Peritoneal Dialysis Hemodialysis Kidney Transplant

Extensive extrarenal No limitation No limitation except Excluded in cardiovasculardisease for hypotension insufficiency

Geriatric patients No limitation No limitation Arbitrary exclusion asdetermined by program

Complete Rare, if ever Very few individuals Common as long rehabilitation as graft functions

Death rate Much higher than Much higher than About the same as for nondiabetics for nondiabetics for nondiabetics

First-year survival About 75% About 75% >90%

Survival to second Almost never Fewer than 5% About 1 in 5decade

Progression of Usual and unremitting. Usual and unremitting. Interdicted by functioningcomplications Hyperglycemia and May benefit from pancreas and kidney. Partially

hyperlipidemia metabolic control. ameliorated by correction ofaccentuated. azotemia.

Disadvantages Peritonitis, Blood access a hazard Cosmetic disfigurement,hyperinsulinemia, for clotting, hemorrhage, hypertension, personal expensehyperglycemia, and infection. Cyclical for cytotoxic drugs. Inducedhyperlipidemia. Long hypotension, weakness. malignancy. Humanhours of treatment. More Aluminum toxicity, immunodeficiency virusdays hospitalized than amyloidosis. transmission.either hemodialysis ortransplant.

Patient acceptance Variable, usually Variable, often Enthusiastic during periods ofcompliant with passive noncompliant with good renal allograft function.tolerance for regimen. dietary, metabolic, or Exalted when pancreas

antihypertensive proffers euglycemia.component of regimen.

Bias in comparison Delivered as first choice Treatment by default. All kidney transplant programsby enthusiasts, although Often complicated by preselect those patients withemerging evidence inattention to fewest complications.indicates substantially progressive cardiac and Exclusion of older patients higher mortality than peripheral vascular obviously favorablyfor hemodialysis. disease. prejudices outcome.

Relative cost Most expensive over Less expensive than Pancreas/kidney engraftmentlong run. kidney transplant in first most expensive uremia therapy

year, subsequent years for diabetic. After first year, more expensive. kidney transplant alone is

lowest-cost option.

(From Friedman,11 1999).

Peritoneal dialysisIn the United States, about 7% of diabetic ESRD

patients are sustained by a form of PD, continuousambulatory peritoneal dialysis (CAPD).9 The advan-tages of this form, include freedom from a machine,ability to perform the procedure at home, rapid train-ing, minimal cardiovascular stress, and avoidance ofheparin.11 Mechanical cycling of dialysate, termed con-tinuous cyclic peritoneal dialysis, can be performedduring sleep.11 Some clinicians characterize CAPD as a“first-choice treatment” for diabetic ESRD patients.

Not all reports on CAPD are positive, however. Rubinand Hsu13 offered a less-enthusiastic assessment ofCAPD in diabetic patients based on experience in alargely African American diabetic population inJackson, Miss. Only 34% of patients remained onCAPD after 2 years, and at 3 years, only 18% continued on CAPD.

A decision to select CAPD must therefore weigh itsbenefits (including freedom from a machine and elec-trical outlets and ease of travel) against its disadvan-tages (unremitting attention to fluid exchange, constantrisk of peritonitis, and disappearing exchange surface)for each patient.11

Kidney transplantationFollowing a kidney transplant, patient survival at

1 and 2 years in diabetic recipients is now similar tothat of the overall ESRD transplant population.9

Statistical superiority in survival after a kidney trans-plant compared with dialytic therapy does not tell thewhole story (Figure 1), however, as rehabilitation isalso vastly improved after transplant. Enhanced qualityof life (QOL) makes kidney transplant the distinctlyfavored treatment when presented to newly evaluateddiabetic ESRD patients younger than age 60 (Table 2).More than half of diabetic kidney transplant recipientsin most series live for at least 3 years, and many survivors return to occupational, school, and homeresponsibilities.11

Unfortunately, bias in assignment to a specific treat-ment may have prejudiced the favorable view of kid-ney transplants. For example, studies in which themean age of transplant patients is a decade youngerthan the CAPD or HD groups are likely to discern bet-ter functional status in the younger group.11

Evaluating the option of a combined pancreas andkidney transplant for the diabetic ESRD patient is diffi-cult. Although still regarded as investigational by some,pancreas transplantation is growing in acceptabilityand technical success.14, 15 A recent American DiabetesAssociation position statement recommends that sim-ultaneous pancreas-kidney (SPK) and pancreas-after-kidney transplantation should be routine in diabetickidney transplant recipients.15

Comorbid risk factorsManagement of diabetic patients with progressive

chronic kidney disease (CKD) is more difficult than inage- and gender-matched nondiabetic patients.Treatment is often complicated by comorbid conditionsaffecting other organ systems, including the eyes(retinopathy), heart (cardiovascular disease), and nervous system (peripheral or autonomic neuropathy).Associated clinical syndromes must be followed andtreated, if possible, while preparing the patient toreceive RRT.16

The toll of coincident disease—especially blindness,limb amputations, and cardiac disease—complicatesrehabilitation. Diabetic ESRD patients have a higherdeath rate due to cardiac decompensation, stroke, sepsis, and pulmonary disease than do nondiabeticESRD patients.11

Table 3 lists the major comorbid conditions of con-cern in the management of diabetic ESRD patients.Retinal, cardiovascular, and lower-limb diseases rank at the top of the list. We routinely collaborate with anophthalmologist so that laser and/or vitreous surgery

Table 3. Diabetic Complications That Persist and/or Progress During ESRD

1. Retinopathy, glaucoma, cataracts

2. Coronary artery disease, cardiomyopathy

3. Cerebrovascular disease

4. Peripheral vascular disease: limb amputation

5. Motor neuropathy, sensory neuropathy

6. Autonomic dysfunction: diarrhea,

dysfunction, hypotension

7. Myopathy

8. Depression

(From Friedman,11 1999).

18

19

are integrated into the comprehensive management ofthese patients. Similarly, we consult—even in asympto-matic patients—a cardiologist familiar with uremia indiabetic patients. Coronary angiography (if indicated) isa valuable tool for detecting patients for whom prophy-lactic coronary artery angioplasty or bypass surgery islikely to extend life. A podiatrist on our team deliversroutine foot care. By regular surveillance of patients atrisk of major lower-extremity disease, the podiatrist hassharply reduced the chance of amputations, a compli-cation we noted in about 20% of patients who do notreceive podiatric care.

Pre-ESRD interventionThe Diabetes Control and Complications Trial

found that “intensive diabetes treatment”—designed toachieve blood glucose levels as close to normal as pos-sible—slows the progression of nephropathy, retinopa-thy, and neuropathy in diabetic patients.17 Several otherinterventions have also been found to be effective indecreasing the rate of development of nephropathy indiabetic patients, including blood pressure reductionand dietary protein restriction. Treatment with antihy-pertensive agents (especially angiotensin-convertingenzyme inhibitors), even in normotensive diabeticpatients, can decrease microalbuminuria, postpone theevolution from microalbuminuria to overt proteinuria(macroalbuminuria), decrease protein excretion inpatients with overt proteinuria, and slow the anticipateddecline in the glomerular filtration rate (GFR).18 Dietaryprotein restriction can decrease microalbuminuria,overt proteinuria, and the anticipated rate of decline in the GFR.18

CKD induces mild anemia that becomes moresevere as kidney disease progresses, no matter whatthe primary cause of kidney functional loss. In diabeticnephropathy, the onset of anemia caused by erythro-poietin deficiency can occur early, during the period of chronic renal insufficiency (CRI).19 A 1999 reportshowed that only 23% of all patients who begin dialysiswith a hematocrit level of <28% received treatmentwith recombinant human erythropoietin (rHuEPO)before reaching ESRD.20

Anemia affects the patient with diabetes and CKD ina number of ways. A lower hemoglobin level reducesoxygen carriage, which in turn impinges on variousaspects of QOL, including the ability to do work, thinkclearly, and enjoy life.21 This decrease in QOL may bereversed with rHuEPO treatment.21 It has been con-

vincingly demonstrated that treatment of anemia withrHuEPO reduces morbidity and mortality in patientswith ESRD who are undergoing dialysis.22 Correction ofanemia, therefore, has become a key component inpreserving work, school, and home responsibilities inpatients afflicted with CKD in diabetes. A well-designed, randomized, prospective trial will help toconvince physicians that treating anemia earlier canreduce the complications and perhaps even delay theprogression of CKD.

Future directions of therapySurvival rates are continually improving for diabetic

ESRD patients on dialytic therapy or after kidney trans-plantation. Progress in developing new therapiesreflects multiple small advances in understanding thepathogenesis of extrarenal micro- and macrovasculopa-thy in diabetes, coupled with safer immunosuppres-sion. Trials based on prevention of tissue and organdamage resulting from hyperglycemia and oxidativestress are now in progress at multiple centers. As clari-fication of the perturbed biochemical reactions under-lying the pathogenesis of diabetic vasculopathy pro-ceeds, we believe a way will be found to block end-organ damage without correcting hyperglycemia,meaning that diabetic nephropathy, like other vasculo-pathic complications, may be entirely preventable.

Accelerated synthesis and tissue deposition ofadvanced glycosylation endproducts (AGEs) has beenproposed as a contributing mechanism in the patho-genesis of clinical complications in diabetes23 and inthe complications of CKD.24 AGEs are the end result of a nonenzymatic reaction called the Maillard reactionthat occurs between ambient glucose and primaryamino groups on proteins.25 This reaction results in glycated residues called Amadori products, which, aftera series of dehydration and fragmentation reactions,are converted into AGEs.

Studies in rodents suggest that AGEs exert their toxicity by impairing nitric oxide-mediated processes,including neurotransmission and blood flow in smallvessels.26 Aminoguanidine, an inhibitor of nonenzym-atic glycosylation and nitric oxide quenching, mayoffer a means for reducing the formation and toxicityof AGEs.26, 27 Separate multicenter trials in types 1 and 2 diabetics whose proteinuria is attributable to nephro-pathy are in progress. This approach is attractivebecause it bypasses the need to achieve euglycemia,which is often unattainable.

20

Alternatively, several exciting and entirely freshapproaches to attaining euglycemia are now being pur-sued. For example, after the function of the insulinreceptor was elucidated, a search was initiated through50,000 compounds to unearth any that might stimulatethe receptor. A compound extracted from the Kenyanfungus Pseudomassaria was discovered to have thecapacity to mimic insulin action in biochemical andcellular assays, acting as an “insulin pill.”28 In twomutant mouse models of diabetes, oral administrationof this compound lowered blood glucose levels by up to 50%, equivalent to the reduction attained with current oral antidiabetic therapies.

The novel concept of cell receptor modification toalter downstream signaling molecules is also beingextended to organ transplant recipients in what hasbeen termed “the new immunology.”29, 30 For individualswith diabetes whose vasculopathy is still best managedby transplant, this technology offers potential promisefor long-term successful “tolerance” of SPK allografts.15

Until that occurs, the ESRD patient with type 1 dia-betes still would be well advised to consider an SPKtransplant as the preferred therapy for curing an inex-orable disease. Within the next few years, however,there is a high probability that either insulin receptorstimulation or restoration of islet function by injectingallogenic or xenogenic islet beta cells will end the “bigsurgery” required for segmental or whole-pancreasgrafting, simplifying or eliminating the stress of surgery while attaining the same objective.15 For thosewith type 2 diabetes, I believe it is likely that an “all-oral” regimen that forestalls vasculopathic complica-tions will be developed.

SummaryThe rate of kidney functional decline in diabetic

nephropathy is slowed by normalization of hyperten-sive blood pressure, establishment of euglycemia, andreduced dietary protein intake. When compared topatients with other causes of ESRD, the diabetic patientsustains greater mortality and morbidity due to con-comitant (comorbid) systemic disorders, especiallycoronary artery and cerebrovascular disease. A func-tioning kidney transplant provides the uremic diabeticpatient better survival with superior rehabilitation thandoes either CAPD or maintenance HD. For some dia-betic ESRD patients, a combined pancreas and kidneytransplant may cure diabetes and permit full rehabilita-

tion. No matter which ESRD therapy has been elected,effective rehabilitation in diabetic ESRD patients is con-tingent on recognition and management of comorbidconditions. There is every reason to believe that the“end of diabetes” as an important clinical disorder willsoon be upon us. Just as poliomyelitis yielded to vac-cines and now is presented as an historical disease tomedical students, so will all of the terrible complica-tions of diabetes be a footnote to a bygone era.

References1. King H, Aubert R, Herman W. Global burden of diabetes, 1995-2025:

prevalence, numerical estimates, and projections. Diabetes Care.1998;21:1414-1431.

2. National Institutes of Health. National Diabetes InformationClearinghouse, Diabetes Statistics. Bethesda, Md: National Instituteof Diabetes and Digestive and Kidney Diseases, NIH Publication No.99-3892, March;1999.

3. Ismail N, Becker B, Strzelczyk P, Ritz E. Renal disease and hyperten-sion in non-insulin-dependent diabetes mellitus. Kidney Int.1999;55:1-28.

4. US Renal Data System. USRDS 2001 annual data report. NationalInstitutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases. 2001.

5. Hasslacher C, Ritz E, Wahl P, Michael C. Similar risks of nephropa-thy in patients with type I or type II diabetes mellitus. Nephrol DialTransplant. 1989;4:859-863.

6. Biesenback G, Janko O, Zazgornik J. Similar rate of progression inthe predialysis phase in type I and type II diabetes mellitus.Nephrol Dial Transplant. 1994;9:1097-1102.

7. US Renal Data System. USRDS 1997 annual data report. Bethesda,Md: National Institutes of Health, National Institutes of Diabetesand Digestive and Kidney Disease;1997.

8. Friedman E. Diabetic Nephropathy. In: Schrier RW, ed. Atlas ofDiseases of the Kidney. Denver: Blackwell Science;1999:1-20.

9. US Renal Data System. USRDS 2000 annual data report. NationalInstitutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases. 2000.

10. Abourizk N, Dunn J. Types of diabetes according to NationalDiabetes Data Group Classification:limited applicability and need to revisit. Diabetes Care. 1990;13:1120-1122.

11. Friedman E. Diabetic nephropathy: fresh perspectives. FactaUniversitatis, Medicine and Biology Series (Yugoslavia). 1999;6:31-47.

12. Ifudu O, Paul H, Mayers J, et al. Pervasive failed rehabilitation incenter-based maintenance hemodialysis patients. Am J Kidney Dis.1994;23:394-400.

13. Rubin J, Hsu H. Continuous ambulatory peritoneal dialysis: tenyears at one facility. Am J Kidney Dis. 1991;17:165-169.

14. Sollinger H, Ploeg R, Eckhoff D, et al. Two hundred consecutivesimultaneous pancreas-kidney transplants with bladder drainage.Surgery. 1993;114:736-743.

15. Sutherland DE, Gruessner RW, Dunn DL, et al. Lessons learnedfrom more than 1,000 pancreas transplants at a single institution.Ann Surg. 2001;233:463-501.

16. Markell MS, Friedman EA. Diabetic nephropathy. Management ofthe end-stage patient. Diabetes Care. 1992;15:1226-1238.

21

17. Diabetes Control and Complications Research Group D. Effect ofintensive therapy on the development and progression of diabeticnephropathy in the Diabetes Control and Complications Trial. KidneyInt. 1995;47:1703-1720.

18. Nathan DM. Long-term complications of diabetes mellitus. N Engl JMed. 1993;328:1676-1685.

19. Bosman DR, Winkler AS, Marsden JT, et al. Anemia with erythro-poietin deficiency occurs early in diabetic nephropathy. Diabetes Care.2001;24:495-499.

20. Obrador GT, Ruthazer R, Arora P, Kausz AT, Pereira BJ. Prevalenceof and factors associated with suboptimal care before initiation of dialy-sis in the United States. J Am Soc Nephrol. 1999;10:1793-1800.

21. Revicki D, Brown R, Feeny D, et al. Health-related quality of lifeassociated with recombinant human erythropoietin therapy for predial-ysis chronic renal disease patients. Am J Kidney Dis. 1995;25:548-554.

22. Harnett JD, Kent GM, Foley RN, Parfrey PS. Cardiac function andhematocrit level. Am J Kidney Dis. 1995;25:S3-7.

23. Brownlee M, Cerami A, Vlassara H. Advanced glycosylation endproducts in tissue and the biochemical basis of diabetic complications.N Engl J Med. 1988;318:1315-1321.

24. Aso Y, Inukai T, Tayama K, Takemura Y. Serum concentrations ofadvanced glycation endproducts are associated with the developmentof atherosclerosis as well as diabetic microangiopathy in patients withtype 2 diabetes. Acta Diabetol. 2000;37:87-92.

25. Wautier JL, Guillausseau PJ. Advanced glycation end products, theirreceptors and diabetic angiopathy. Diabetes Metab. 2001;27:535-542.

26. Bucala R, Tracey K, Ceram A. Advanced glycosylation productsquench nitric oxide and mediate defective endothelium-dependentvasodilation in experimental diabetes. J Clin Invest. 1991;87:432-438.

27. Edelstein D, Brownlee M. Mechanistic studies of advanced glycosy-lation end product inhibition by aminoguanidine. Diabetes. 1992;41:26-29.

28. Zhang B, Salituro G, Szalkowski D, et al. Discovery of a small mole-cule insulin mimetic with antidiabetic activity in mice. Science. 1999;284:974-977.

29. Schwartz R. The new immunology - the end of immunosuppres-sive drug therapy? N Engl J Med. 1999;340:754-755.

30. Guinan EC, Boussiotis VA, Neuberg D, et al. Transplantation ofanergic histoincompatible bone marrow allografts. N Engl J Med.1999;340:1704-1714.

22

lit review

Review and commentary on the recent literatureArticle title: Remission and regression in the nephropathy of type 1 diabetes when

blood pressure is controlled aggressively

Authors: Hovind P, Rossing P, Tarnow L, Smidt UM, Parving HH

Journal: Kidney Int. 2001;60:277-283

lit review

here is a growing body of literature on the nephro-pathy associated with diabetesmellitus, encompassing natural historyand interventional studies, which suggests

that treatment of hypertension may result in reductionin proteinuria and rate of loss of kidney functionamong patients with diabetic and nondiabetic kidneydisease. This observational study by Hovind and colleagues, of 301 patients with nephropathy due totype 1 diabetes, further contributes to this body of work.

The authors studied consecutive patients at a singlecenterwho had a minimum of 3 years of follow-up,with yearly assessment of albuminuria and glomerularfiltration rate (GFR) using a 51Cr-EDTA plasma clear-ance technique. Remission was defined as a decreasein albuminuria to <200 mg/min and ≥30% reduction,sustained for at least 1 year. Regression was defined as a decline of GFR <1 mL/min/year during the observation period.

The mean rate of decline of GFR was 4.0 ±0.2 mL/min/year. There were 30 patients (10%) who were spontaneously normotensive throughout thestudy. Sixty-seven patients (22%) experienced regression and 92 (31%) entered remission.

Systolic and diastolic blood pressures were signifi-cantly lower among patients with regression or remis-sion compared to those without. Among patients in theregression group, 49% obtained remission, while 36%in the remission group achieved regression. From thepersistent normotensive group (N=30), spontaneousremission and regression was noted in 10 and 14patients, respectively, of whom none received antihy-pertensive treament. Logistic regression analysisshowed an adjusted odds ratio of 2.14 (95% CI, 1.33 to3.44) for regression for every 10 mm Hg lower bloodpressure, as well as a higher odds for regression withlower albuminuria and better glycemic control.

From these results, the authors conclude that“aggressive antihypertensive treatment can induceremission and regression in a sizable fraction of type 1diabetic patients,” and that “genuine normotensive...patients with diabetic nephropathy have an excellent kidney prognosis.”

CommentaryThis cohort study confirms prior observational stud-

ies of the impact of various factors on the progression

of diabetic kidney disease, with the dual strengths ofusing a reliable method for ascertaining GFR, and having a fairly large sample size. Larger interventionalstudies (such as the Modification of Diet in RenalDisease1 and Appropriate Blood Pressure Control inDiabetes2 trials) have confirmed the benefit of bloodpressure lowering among patients with nondiabeticand diabetic kidney disease.

The mean baseline GFR of the study group was fairly high, and it is unclear whether patients who pro-gressed to end-stage renal disease within 3 years wereincluded in the study. If these patients were excluded,the observed mean rate of decline of GFR is lower thanwould be expected if all patients were included. Thus,the study would have assessed the impact of bloodpressure and other factors in the lower-risk group of slow progressors or nonprogressors.

The authors’ conclusion that aggressive treatment ofhypertension results in remission and regression maybe too strongly stated, as there was a sizeable propor-tion of patients (N=30) who were spontaneously normotensive in the regression group. Overall, thisgroup had a significantly lower rate of decline of GFR (1.9 mL/min) compared with

the hypertensive patients requiring antihyperten-sive treatment (4.3 mL/min). In addition, only 40% of patients achieved adequate blood pressure control(<140/90 mm Hg). It could be argued that the benefitobserved in the study possibly comes from remainingnormotensive, not from adequate treatment of hyper-tension. It would have been of interest to evaluate theinteraction between blood pressure and use of anti-hypertensive agents.

Nevertheless, this is a well-performed study utilizinga reliable and reproducible measurement of kidneyfunction, which confirms the potential benefit of managing hypertension and diabetes even in earlier stages of kidney functional impairment.

Annamaria T. Kausz, MDBoston, Massachusetts

References

1. Hunsicker LG, Adler S, Caggiula A, et al. Predictors of the progres-sion of renal disease in the Modification of Diet in Renal DiseaseStudy. Kidney Int. 1997;51:1908-1919.

2. Estacio RO, Jeffers BW, Gifford N, Schrier RW. Effect of blood pressure control on diabetic microvascular complications in patientswith hypertension and type 2 diabetes. Diabetes Care. 2000;23(suppl 2):B54-B64.

T

23

Article title: Effects of ramipril vs amlodipine on renal out comes in hypertensive nephrosclerosis: a randomized controlled trial

Authors: Agodoa LY, Appel L, Bakris GL, et al

Journal: JAMA. 2001;285:2719-2728

he results of this study, from the AfricanAmerican Study of Kidney Disease andHypertension (AASK) Study Group, sug-gest that antihypertensive therapy with

ramipril offers greater benefit in slowing deteriorationof kidney function than does amlodipine in patientswith hypertensive kidney disease and proteinuria.Ramipril is an angiotensin-converting enzyme (ACE)inhibitor; amlodipine is a dihydropyridine calciumchannel blocker (DHP-CCB). The study also suggestedthat ramipril may offer benefit to patients without pro-teinuria.

The AASK trial was designed to evaluate the impactof each of three different classes of antihypertensivedrugs—ACE inhibitors, DHP-CCBs, and beta-blockers—on the progression of hypertensive kidney disease inAfrican Americans. The patients (n=1,094), ranging inage from 18 to 70 years, had glomerular filtration rates(GFRs) of 20 to 65 mL/min/1.73 m2 and were enrolledin the study between February 1995 and September1998, with planned follow-up through September 2001.

The authors performed subgroup analyses on partici-pants with baseline urinary protein-to-creatinine ratio(UP/Cr) of >0.22 and ≤0.22, a value correspondingapproximately to the threshold of 300 mg/d for clinical-ly significant proteinuria. The UP/Cr cutpoint of 0.22was chosen post hoc, but was selected because of clini-cal relevance independent of the AASK data, accordingto the authors.

At follow-up, among participants with a UP/Cr of>0.22, the ramipril group had a 36% slower meandecline in GFR over 3 years and a 48% reduced risk ofclinical endpoints compared to the amlodipine group.Among the entire cohort, there was no significant dif-ference in mean GFR decline from baseline to 3 yearsamong treatment groups. However, the ramipril grouphad a 38% reduced risk of clinical endpoints and a 36%slower mean decline in GFR after 3 months comparedto the amlodipine group.

The authors note that this study is the first clinicalendpoint trial with sufficient sample size to evaluatethe effect of inhibition of the renin-angiotensin-aldos-terone system on kidney disease in African Americans.The results support the use of ramipril as initial therapy in a multidrug regimen over a DHP-CCB-basedregimen in patients with mild to moderate chronic kidney disease (CKD) and proteinuria.

In September 2000, the amlodipine arm of the AASKtrial was terminated on the recommendation of thetrial’s data and safety monitoring board, based onresults related to the treatment interventions with baseline proteinuria. External clinical studies releaseddata, after the initiation of AASK, suggesting that thereis a slowing of the progression of renal disease by ACEinhibitors in participants with elevated proteinuria andalso suggesting that DHP-CCBs may elevate the level of proteinuria in patients and not slow the progressionof renal disease.

CommentaryA potential drawback to this study is that it used

only half the number of patients in the amlodipinearm, based on a pilot study showing a short-termincrease in GFR. One could argue that too muchemphasis was placed on the results of the pilotstudy, which included a total of only 94 patients.1

In addition, the decision to stratify patients into subgroups according to their baseline UP/Cr was made post hoc, making it seem somewhat arbitrary.

The primary outcome measured in this study wasthe change in GFR. By combining patients with UP/Crabove and below 0.22, however, the authors fail to provide convincing evidence that either agent is supe-rior in slowing the decline in GFR. This demonstrateshow introducing artificial subdivisions after the factcan lead to problems in analyzing results from a ran-domized controlled trial. According to the data, itappears that amlodipine is better if UP/Cr <0.22,worse if UP/Cr >0.22, better if GFR >40 mL/min, andworse if GFR <40 mL/min. It would be difficult to pro-vide a biological explanation for the latter statement.

What is convincing, however, is the disparity in“hard” events between the two groups. The amlodipinegroup clearly fares worse than the ramipril group withregard to death and end-stage renal disease. The studylacks a placebo group, so that it is a matter of specu-lation whether this is because ramipril imparts benefit,amlodipine imparts risk, or both. At the very least, thestudy confirms the belief that ACE inhibitors should bethe antihypertensive agents of choice in CKD patientswith hypertension, and, importantly, confirms that thisis also the case among African Americans.

Robert N. Foley, MDSalford, England

References

1. Hall WD, Kusek JW, Kirk KA, et al. Short-term effects of blood pressure control and antihypertensive drug regimen on glomerularfiltration rate: the African American Study of Kidney Disease andHypertension Pilot Study. Am J Kidney Dis. 1997;29:720-728.

T

24

Scarring may play major role in transplant rejection

One of the body’s own mechanisms for wound healing might be a major contributor to chronic kidneyrejection following transplantation, according to astudy1 published in the New England Journal of Medicine.

Tissue remodeling is known to be a prominent feature of chronic kidney transplant rejection. Mesen-chymal cells (fibroblasts and myofibroblasts) are keyplayers in this process. However, no one knowswhether these cells originate locally or from precursorcells from the recipient’s own circulation. Using DNAanalysis to study biopsies from transplanted kidneysthat were undergoing chronic rejection, the authorsfound that up to 38% of the mesenchymal cells in theallografts originated from the recipient rather than thedonor. Donor mesenchymal cells were also present.

The authors conclude that the presence of mes-enchymal cells of host origin in the vascular and inter-stitial compartments of kidney allografts undergoingchronic rejection provides evidence that a circulatingmesenchymal precursor cell has the potential tomigrate to areas of inflammation.

Compliance not an issue in treatment-resistant hypertension

For many patients with arterial hypertension, bloodpressure cannot be adequately controlled, despite treat-ment with antihypertensive drugs. Noncompliancewith treatment is thought to be common in patientswith treatment-resistant hypertension. However, aSwiss study2 reported in the British Medical Journal hasfound no difference in compliance between treatment-resistant and treatment-responsive patients.

According to one definition, treatment-resistanthypertension is persistent high blood pressure(>140/90 mm Hg for patients aged ≤60 years or>160/90 mm Hg for those aged >60 years) in spite of treatment for a sufficient duration with at least twoappropriate antihypertensive drugs. Various explana-tions have been given for treatment-resistant hyperten-sion, including secondary hypertension, endogenousresistance to treatment, and, foremost, noncompliancewith antihypertensive drug regimens.

To test how much of a role noncompliance actuallyplays in resistant hypertension, the researchers studied103 patients who had been taking between two andfour drugs for at least 1 month. Using electronic pill

boxes that record every opening, they monitoredwhether these patients took their medication over a period of 4 weeks. Patients who took 80% or more of their prescribed doses were considered to be compliant.

The study found no difference in compliancebetween treatment-resistant and treatment-responsivepatients. Forty (82%) of the 49 treatment-resistantpatients were compliant, while 46 (85%) of the 54 treatment-responsive patients were compliant.

The authors conclude that other factors independentof a patient’s willingness to adhere to a treatment regi-men are more relevant in explaining treatment resist-ance in most patients, and that these factors should beexamined to explain lack of response to antihyper-tensive drugs.

Unexpected treatment for type 1 diabetes: TNF-α

Researchers at Harvard Medical School have foundan unexpectedly simple treatment for type 1 diabetesin mice. In a study3 published in the Journal of ClinicalInvestigation, the researchers were able to retrain theimmune system in order to halt the autoimmuneprocess that causes the destruction of islet cells innonobese diabetic (NOD) mice.

The study presents two key findings. First, theautoimmune cells in NOD mice are unusually sensitiveto the cytokine tumor necrosis factor-α (TNF-α), whichcauses them to undergo apoptosis (programmed celldeath). Second, many of these same cells are unable to present self-antigens, a process that is crucial for pre-venting the development of autoimmune reactions.

The Harvard team thus designed a two-prongedtreatment strategy. First, they triggered the expressionof TNF-α in the NOD mice in an attempt to destroy the immune cells that had gone awry. They then trans-planted spleen or pancreas cells from non-autoimmunemice in order to expose these mice to normal majorhistocompatibility complex class I (MHC-I) self-antigens.

Ordinarily, transplants of insulin-producing pancre-atic islet cells from healthy donors do not survive forvery long in either NOD mice or humans. But the TNF-α/MHC-I treatment not only reversed islet-cellautoimmunity in the NOD mice, it also restored endo-genous pancreatic islet function to such an extent thatnormal blood glucose levels were maintained in up to

what’s newwhat’s new

25

75% of animals after the treatment was discontinuedand islet transplants were removed.

The authors conclude that it might be feasible toapply this treatment regimen to humans with type 1diabetes as well.

New interactive CD-ROM available forchildren with kidney disease

The Starbright Foundation has created a new CD-ROM titled "Living with Kidney Disease," for childrenand teenagers with kidney disease. This multimediaprogram offers information on everything from med-ications, diet, and dialysis to tips from peers and whatto expect before and after a transplant.

The new, interactive CD-ROM is part of theStarbright Explorer Series. The series also includes CD-ROMs for other serious medical conditions, such as cystic fibrosis and sickle cell anemia.

The Starbright Foundation is a nonprofit organ-ization, chaired by Steven Spielberg and NormanSchwarzkopf, that creates programs designed toempower children with serious medical conditions to address the psychosocial challenges that accompanytheir illnesses.

More information on the Starbright Explorer Series isavailable at www.starbright.org.

Cocaine use, hypertension, and end-stage renal disease

Cocaine use has been associated with both acute kid-ney failure and hypertension (HTN), but recent dataalso suggest it may lead to a chronic insidious form ofkidney failure. In a recent cross-sectional study4 ofurban patients who were undergoing outpatienthemodialysis, researchers from the University ofCalifornia, Los Angeles, compared the rates of cocaineuse in patients with and without HTN-related end-stage renal disease (HTN-ESRD).

The researchers found that 89% of the cocaine usersenrolled in the study had been diagnosed with HTN-ESRD compared with only 46% of those who did notreport cocaine use. Of the 113 subjects with HTN-ESRD, 43% had a history of cocaine abuse. Those HTN-ESRD patients who reported cocaine use were also significantly younger than those who had not usedcocaine, and had also been hypertensive for a shorterperiod of time before they needed dialysis.

Based on these results, the authors conclude thatcocaine should be considered as a cause of ESRD inpatients without a clear cause of kidney failure.

Oral bacterial supplements may reducestone formation

High levels of oxalate in the urine are a major riskfactor for kidney stones, and foods that contain thischemical appear to exacerbate the problem. Increasedintestinal absorption of oxalate also may play a role.Because some normal intestinal bacteria are known tometabolize oxalate in the gut, which in turn limits itsabsorption, Italian researchers recently conducted astudy5 to see whether lactic acid bacteria might helpreduce oxaluria in patients who had formed calcium-oxalate stones.

Six patients were fed a freeze-dried mixture contain-ing the bacterial strains Lactobacillus acidophilus, L plan-tarum, L brevis, Streptococcus thermophilus andBifidobacterium infantis for 4 weeks. At the end of thattime, urinary oxalate excretion was measured. Theteam also studied the ability of each of the bacterialstrains to degrade oxalate and grow in oxalate-contain-ing media.

The researchers found that oxalate excretion wasreduced in all six of the patients, with mean levelsfalling from 55.5 mg/d to 33.5 mg/d. They concludethat treatment with a high concentration of freeze-dried lactic acid bacteria can greatly reduce the urinary excretion of oxalate, and that this biologicalmanipulation of endogenous digestive microflora maybe a novel approach for the prevention of urinary stone formation. CKD

References

1. Grimm P, Nickerson P, Jeffery J, Savani RC. Neointimal and tubu-lointerstitial infiltration by recipient mesenchymal cells in chronicrenal-allograft rejection. N Engl J Med. 2001;345:93-97.

2. Nuesch R, Schroeder K, Dieterle T, Martina B, Battegay E. Relationbetween insufficient response to antihypertensive treatment andpoor compliance with treatment: a prospective case-control study.BMJ. 2001;323:142-146.

3. Ryu S, Kodama S, Ryu K, Schoenfeld DA, Faustman DL. Reversal ofestablished autoimmune diabetes by restoration of endogenous betacell function. J Clin Invest. 2001;108:63-72.

4. Norris KC, Thornhill-Joynes M, Robinson C, et al. Cocaine use,hypertension, and end-stage renal disease. Am J Kidney Dis.2001;38:523-528.

5. Campieri C, Campieri M, Bertuzzi V, Swennen E, Matteuzzi D.Reduction of oxaluria after an oral course of lactic acid bacteria athigh concentration. Kidney Int. 2001;60:1097-1105.

26

Upcoming Meetings

FebruaryFebruary 7 — 9

Title: 27th International Stroke Conference. Sponsor: Stroke Council of theAmerican Heart Association. Location: Henry B. Gonzalez Convention Center,

San Antonio, Texas. Website: http://www.strokeconference.org/index.oft

February 28 — March 2Title: International Society of Renal Nutrition and Metabolism in Renal Disease:

11th International Congress on Nutrition and Metabolism in Renal DiseaseLocation: Nagoya, Japan. E-mail: [email protected]

MarchMarch 2 — 6

Title: Annual Conference on Dialysis. Location: Tampa, Florida Phone: (573) 882-4105. Website: http://muextension.missouri.edu

March 23 — 26Title: Renal Physicians Association/Research and Education Foundation (RPA/REF)

Annual Meeting. Location: Washington, DC. E-mail: [email protected]

AprilApril 22 — 26

Title: The Genomics Revolution: Bench to Bedside to Community, and the 42nd Annual Conference on Cardiovascular Disease Epidemiology and

Prevention. Location: Hawaii Convention Center, Honolulu, Hawaii. Website: http://www.americanheart.org/asianpacific/index.html

April 26 — May 3Title: 3rd Annual Joint Meeting and Scientific Sessions, American Society of Transplantation/American Society of Transplant Surgeons (AST/ASTS).

Location: Washington, DC. E-mail: [email protected]

27

MayMay 23 — 26

Title: 33rd National Symposium, American Nephrology Nurses’ Association(ANNA). Location: Orlando, Florida. Website: http://www.annanurse.org

JuneJune 2 — 5

Title: 13th World Congress. Sponsor: International Society for the Study of Hypertension in Pregnancy (ISSHP). Location: Toronto, Canada.

E-mail: [email protected]

JulyJuly 14 — 17

Title: 2002 Congress of the European Renal Association (ERA)/European Dialysisand Transplant Association (EDTA). Location: Copenhagen, Denmark.

Website: http://www.unipr.it/~eraedta/

Treatment of hypertensionYour Opinion presents the results of a reader survey that welcomes the participation of all nephrologists (see next issue’s survey below). For this issue, a randomly selected group of nephrologists were asked about hypertension and sexual dysfunction as they relate to renal disease. Here are their aggregate responses.

0% 0%

5%

0% 0% 0% 0%

<10%

% o

f Nep

hrol

ogis

ts s

urve

yed

% of Patients

10-25% 51-75% >75%

CKD

21%

68%

100%

26-50%

ESRD

100%

120%

80%

60%

40%

20%

20%

16%5%

0% 0% 0%0%0%

42%

26%

<10%

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0

% o

f Nep

hrol

ogis

ts s

urve

yed

% of Patients

10-25% 51-75% >75%

CKD

52%

74%

26-50%

ESRD

11%

6%

22%

56%

6% 6%

<10%

0.6

0.5

0.4

0.3

0.2

0.1

0

% o

f Nep

hrol

ogis

ts s

urve

yed

% of Patients

10-25% 51-75 >75%

CKD

33%

22%

28%

11%

26-50%

ESRD

0%

1

Choices ranked from 1 to 6

2

3

4

5

6

20% 40% 60% 80% 100%

% of Nephrologists surveyed

ACEIs

Park

ing ARBs

Non-dihydropyridine CCBs

Dihydropyridine CCBs

Beta-blockers

Diuretics

your opinionyour opinion

What percentage of your patients have diabetesmellitus (DM) as a cause of renal disease?

What percent of your patients have DMwhether or not it is the cause of renal disease?

What percent of your patients with diabeteshave a regular endocrinologist who participates in their care?

Rank your preference of antihypertensive class in diabetic patients with chronic kidney disease (CKD):

28

Comment for publication

The nephrologists in this survey were asked: Whatare the barriers to nephrologists providing the principalcare related to diabetes for end-stage renal disease orCKD patients? Some of the responses included:

Limited resources in terms of time, patient load,and the number of parameters that need to be man-aged per patient

Lack of a system for coordinating care betweennephrologist, diabetes clinic and supportstaffLimitations imposed by managed care

Turf battles

Poor referral rate, lack of access to these patients

Inability to track the detailed management of thesepatients, which needs to be multi-disciplinary

Note—"Your Opinion" is an informal poll that makes noclaims of statistical significance

We want your opinion for our next issue!

Fax or mail us your responses to the following questionsregarding your patients with diabetes mellitus (DM):

1. Do you have an educational program for newpatients (check one)?

Yes ____ No ____

If you answered "Yes" above, please check the items offered in your program, and the approximate pecentage of new patients who receive them.

Check all that apply Approximate % of new patients

Written material ____ %Videotapes ____ %Direct teaching by physician ____ %

Direct teaching by nurse educator ____ %

Other _ ____ %

2. Please rank each of these challenges 1 to 4 in termsof how large an obstacle they are to patient education(1 = significant obstacle; 4 = no obstacle).

Circle the appropriate number in each row

(1 = significant obstacle; 4 = no obstacle):

Cultural/language barriers 1 2 3 4

Lack of interest/ability to focus by patients 1 2 3 4Lack of adequate teaching materials 1 2 3 4Lack of time by dialysis staff 1 2 3 4

Your comment for publicationWhat are the barriers to nephrologists providing the principal care related to diabetes for ESRD or CKD patients?

Note—“Your Opinion” is an informal poll that makes no claims of statistical significance.

Name:____________________________Phone: ____________City:__________________________State: ______________

Fax the completed survey to 973-263-3952 or send to: CKD Survey, HealthVizionCommunications, 20 Waterview Blvd., Parsippany, NJ 07054-1295.

29

30

INITIAL VISIT: NOVEMBER 1996

Case description: A 76-year-old white woman pre-sented to the Osteoporosis Clinic for evaluation ofsevere osteoporosis and spontaneous compression frac-tures. She complained of severe back pain, and had a

20-year history of frequent use of over-the-counteranalgesics, including headache powders.

Family history: Negative for renal insufficiency;positive for osteoporosis in multiple family members.

Two to three alcoholic drinks per day; 1 pack of cigarettes per day.

diagnosticchallenge

diagnosticchallenge

Table 1: Physical Examination Test Results

Table 2: Laboratory Test Results

Submitted by

Jill Lindberg, MDNew Orleans, Louisiana

Test November 1996 February 1999

Weight (lb) 104 88

Blood pressure (mm Hg) 120/80 100/70

Peripheral edema None None

Kyphosis + with tenderness + severe without

over T3, T4 tenderness

Pallor _ +

EKG +LVH +LVH with BBB

DEXA standard deviations _3.5 _4.0below young normal at thefemoral neck

Chest x-ray Positive T4-T5 Pulmonary edemacompression fractures No new compression

fractures

Test/Findings November 1996 February 1999

Creatinine (mg/dL) 1.6 2.2

Creatinine clearance (cc/min) 30.0 12.0

Hematocrit (%) 32.0 29.0

Serum calcium (mg/dL) 8.0 7.8

Albumin (g/dL) 3.8 2.9

Phosphorus (mg/dL) 2.0 4.0

PTH (ng/ml) 122.0 380.0

Urine protein (mg/day) 400.0 502.0

Urinalysis Multiple white cell casts,1+protein, 1+heme,

negative red cell casts

WBC Normal with 3.290 Normal with 8.790eosinophils eosinophils

Diagnostic Challenge: What was the most likely cause of thispatientÌs renal insufficiency? What was the most likely cause of her severe left ventricular dilatation and resulting congestive heart failure?

Medications: Advil® (ibuprofen, Whitehall-Robins);headache powders, including BC® Headache Powder(aspirin, salicylamide, and caffeine, GlaxoSmithKline)and Goody’s® Headache Powder (acetaminophen,aspirin, and caffeine, GlaxoSmithKline).

Physical examination: See Table 1. Blood pressurewas 120/80 mm Hg; lungs were clear; heart PMI (pointof maximal intensity) was displaced laterally to ananterior axillary line; extremities had no edema; backhad positive tenderness over T3 and T4.

Lab results: See Table 2. In addition, thoracic spinefilms showed two compression fractures. Renal ultrasound revealed increased echogenicity in both kid-neys with kidneys slightly reduced in size to 8.7 cm bilaterally.

Treatment adjustments: Treatment with subcuta-neous calcitonin, 100 units 3 times per week, was initiated, along with Citracal® Caplets +D (calcium citrate with vitamin D3, Mission Pharmaceuticals). Thepatient was instructed to return within 1 month for fol-low-up of her renal insufficiency and education in theHealthy Start program. She also was asked to return in2 weeks for a kidney biopsy, as all serology tests and

workup of her renal insufficiency were negative. Shewas very upset that the physical exam and history hadfocused on her renal insufficiency.The patient refusedto return for kidney biopsy and follow-up. She was lostto follow-up until her visit in 1999.

FOLLOW-UP: FEBRUARY 1999

Patient returned with increased creatinine levels asseen in Table 2. In addition, she was very anemic andcomplained of severe shortness of breath.

Physical exam: Blood pressure was 100/70 mm Hg;weight had dropped to 80 pounds. Lung exam revealedrales halfway up bilaterally; heart was positive S3; PMIwas displaced laterally into the anterior axillary line.

Test results: Renal ultrasound was repeated, andrevealed increased echogenicity. Kidney size was now 8 cm bilaterally. Urinalysis revealed persistent white cell casts. Echocardiogram revealed an ejectionfraction of 20%.

Outcome: The patient refused renal biopsy. She was initiated on hemodialysis after diuresis was unsuccessful.

Hyponatremia in an AIDS patient in this settingshould prompt consideration of 4 major diagnoses: (1)so-called dilutional hyponatremia due to volume deple-tion; (2) the syndrome of inappropriate secretion ofantidiuretic hormone (SIADH); (3) cerebral salt wast-ing; and (4) adrenal insufficiency. The lower urinarysodium at presentation effectively excludes the latter 2diagnoses. The patient is clearly volume depleted. Hehas orthostatic hypotension and a low jugular venouspressure; this probably accounts for the low urinarysodium and high urinary osmolality at presentation.

However, after several days, the patient, despiteremaining hyponatremic, now has substantial amountsof sodium and chloride in the urine and continued evi-dence of vasosuppressant secretion; that is, his urineosmolality is 410 mOsm/L. His condition can no longerbe explained by severe volume depletion. It is likelythat the volume depletion was superimposed on mildSIADH in this patient. The patient with SIADH is quite

capable of normal sodium retention given an appropri-ate volume stimulus; SIADH is not a salt-wasting disorder.

The hyponatremia improved with saline because theSIADH was mild and perhaps becaused it was coupledwith gastrointestinal water losses. The renal insuffi-ciency (not pursued here for the purposes of this dis-cussion) can make the diagnosis of SIADH problematiclargely because of the uncertainty that is introducedregarding sodium handling by the diseased kidney.However, in this case, there's clear evidence that sodi-um conservation is not impaired by the kidney disease.

Also notable is the relatively low uric acid levelwhich may be seen in both SIADH and cerebral salt-wasting. However, it is not widely appreciated thathypouricemia is also seen with guaifenesin(Robitussin), which is uricosuric. The drug was themost common cause of low serum uric acid in theBoston Collaborative Drug Study.

Click here for the audio answer to this challenging case.

Answer to Diagnostic Challenge in CKD Vol 1, Number 2. Syndrome of inappropriate secretion of antidiuretic hormone. Submitted by Richard Sherman, MD, New Brunswick, New Jersey

Amgen Inc.One Amgen Center DriveThousand Oaks, CA91320-1799

www.RenalAdvances.com

© 2002 Amgen Inc.All rights reserved.

P35028MC1549715M/06-01