American Journal of Kidney Diseases
Volume 48, Issue 1 , Pages 159-166, July 2006

Surrogate Markers in Clinical Studies: Problems Solved or Created?

  • Braden Manns, MD

      Affiliations

    • Departments of Medicine and Community Health Sciences, University of Calgary, Institute of Health Economics, Edmonton, Alberta, Canada
    • ,
  • William F. Owen Jr, MD

      Affiliations

    • Office of the Chancellor, University of Tennessee, Health Sciences Center, Memphis, TN
    • ,
  • Wolfgang C. Winkelmayer, MD, ScD

      Affiliations

    • Division of Pharmacoepidemiology and Pharmacoeconomics, and Renal Division, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA
    • ,
  • P.J. Devereaux, MD

      Affiliations

    • Departments of Medicine and Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, Ontario, Canada
    • ,
  • Marcello Tonelli, MD, SM, FRCPC

      Affiliations

    • Institute of Health Economics, Edmonton, Departments of Medicine and Critical Care, University of Alberta, Alberta, Canada
    • Corresponding Author InformationAddress reprint requests to Marcello Tonelli, MD, SM, FRCPC, University of Alberta, Division of Nephrology and Immunology, 11-103C Clinical Science Bldg, 8440 112 St, Edmonton, Alberta T6B 2B7, Canada

Received 12 December 2005; received in revised form 21 March 2006 published online 09 May 2006.

Article Outline

 

A SURROGATE END POINT is a laboratory measurement or physical sign used as a substitute for a clinically meaningful end point.1 Researchers and pharmaceutical companies commonly assess the efficacy of new agents through evaluation of their impact on surrogate end points in randomized clinical trials (RCTs). This occurs for many reasons, including that regulatory agencies (eg, the US Food and Drug Administration2) can grant marketing approval on the basis of controlled trials showing a favorable effect on a surrogate end point.3 However, very few validated surrogate end points are linked indisputably with the ultimate clinical end points of interest to patients and physicians (ie, survival, need for dialysis, and so on). This article reviews the use of surrogate end points in clinical trials within nephrology and, more broadly, internal medicine, highlighting challenges in interpreting clinical evidence based exclusively on an intervention’s impact on a putative surrogate end point. Given the rapid escalation in costs for health care interventions, we also discuss special challenges that the use of surrogate end points poses for health care payers and clinical practice guideline (CPG) developers, focusing on examples drawn from the literature.

Back to Article Outline

Illustrative Example 

You are seeing a 64-year-old man with end-stage renal disease (ESRD) in your office for follow-up. He has a history of ischemic heart disease, hypertension, and diabetes and has been receiving hemodialysis for the past 18 months. Despite being very compliant with his diet and using calcium carbonate, 500 mg, 3 times daily with meals, and alfacalcidiol, 0.5 μg, orally 3 times weekly, laboratory tests show a serum calcium level of 9.2 mg/dL (2.30 mmol/L), serum phosphate level of 5.9 mg/dL (1.89 mmol/L), and parathyroid hormone (PTH) level of 400 pg/mL (400 ng/L). Having attended a recent nephrology meeting, you wonder whether he would benefit from the use of sevelamer, rather than calcium, and whether the addition of cinacalcet would be beneficial. Your patient, during the ensuing discussion, ponders aloud, “Doctor, I’m willing to take them if they will make me feel better or make me live longer.”

Knowing the importance of this decision, not only for your patient, but also for his health care payer (given an estimated annual cost of US $3,600 for sevelamer and US $8,500 for cinacalcet at usual doses), you retrieve the important clinical trials of sevelamer and cinacalcet. After careful review of the published evidence, you find several clinical trials comparing sevelamer with calcium-based phosphate binders that showed a similar decrease in serum phosphate level, slightly lower (0.4 mg/dL) serum calcium level, and slower rate of progression of coronary calcification scores in patients with ESRD treated with sevelamer.4 You also find several reports suggesting that coronary calcification is a surrogate end point for cardiovascular mortality in patients with ESRD.5, 6

You find several clinical trials showing that cinacalcet can decrease PTH levels in patients with secondary hyperparathyroidism without precipitating hypercalcemia or hyperphosphatemia.7 A review article that you retrieve also suggests that decreasing PTH levels and correcting the biochemical abnormalities of secondary hyperparathyroidism may lead to long-term improvement in bone and vascular pathological states and, possibly, survival.8

Back to Article Outline

Clinical End Points, Clinical Correlates, and Surrogate End Points 

Well-designed and -executed RCTs constitute the gold standard for assessing the efficacy of a new drug. In phase 3 trials, the primary outcome ideally is a clinical end point such as mortality, need for dialysis, or clinical events leading to decreased quality of life. A clinical end point was defined as an outcome that is relevant and noticeable to the patient and therefore is an outcome they wish to avoid9 (Fig 1).

A clinical correlate is a marker that is associated with disease activity, but may or may not change in the same manner as the relevant clinical end point in response to an intervention (Fig 1). For example, in a patient with active bacteremia, acetaminophen may decrease the patient’s fever (potentially a marker of severity of disease), but is unlikely to improve the patient’s probability of survival. In this case, fever is a clinical correlate, but not a surrogate end point.

As noted, a surrogate end point is a laboratory or radiological measurement or physical sign used as a substitute for a clinical end point.1 Surrogate end points represent a subgroup of clinical correlates, and although they are not in themselves important to patients, they are associated with outcomes that are (eg, bone density for fracture, blood pressure for stroke). Unfortunately, nonclinical end points frequently are just clinical correlates, rather than true surrogate end points, and often cannot reliably substitute for clinical end points (Fig 2). In other words, changes in a valid surrogate end point accurately predict changes in a clinical end point in a test of an intervention, whereas clinical correlates (often associated with clinical end points in observational studies) may not modify in the same manner as the clinical end point in response to an intervention.

  • View full-size image.
  • Fig 2. 

    Reasons for failure of surrogate end points. (A) The surrogate is not in the causal pathway of the disease process. (B) Of several causal pathways of disease, the intervention affects only the pathway mediated through the surrogate. (C) The surrogate is not in the pathway of the intervention’s effect or is insensitive to its effect. (D) The intervention has mechanisms of action independent of the disease process. Dotted lines, mechanisms of action that might exist. (Original source: Annals of Internal Medicine9; reprinted with permission from the American College of Physicians.)

Validating a Surrogate End Point 

As noted by Fleming and DeMets,9 “a correlate does not a surrogate make.”9 There are many examples of clinical correlates (some highlighted next) that are not surrogate end points. The process of validating a surrogate end point is complicated,10, 11, 12 statistically controversial,2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 and easily overlooked if a rational pathobiological link is proposed. Generally speaking, validation of a surrogate end point requires a large clinical trial(s) that measures both clinical and surrogate end points. Subsequently, the proportion of treatment effect on the clinical end point explained by a surrogate can be determined by using regression models.13, 14, 15

There are many reasons that potential clinical correlates may fail to satisfy the criteria of a surrogate end point.9 As shown in Fig 2, it often is difficult to differentiate a clinical correlate from a surrogate end point in clinical practice. Moreover, the validity of a surrogate end point is always context dependent; a particular variable may be a valid surrogate end point in certain populations, but not others, and when evaluating certain interventions, but not others. For example, although bone densitometry is considered a useful surrogate end point for fracture reduction using bisphosphonate therapy in patients with osteoporosis, in the next example, we show that bone densitometry is not a surrogate end point for fracture reduction with respect to the use of sodium fluoride in patients with the same disease.16 The context and intervention dependence of surrogate end points serve to highlight the very limited utility of these end points.

Why Use Surrogates? 

Despite their limitations, surrogate end points and sometimes clinical correlates are helpful in various phases in drug development. For example, when determining whether a potential drug candidate is therapeutically active, its effect on a clinical correlate often is measured more easily. Drugs without significant effect generally would not undergo further study. Studies using clinical correlates also may contribute information about clinical pharmacology, including insight into novel indications or the appropriate dosing required to achieve “biological effect” and thus facilitate the design of phase 3 RCTs.17 Given the expense associated with pharmaceutical research, this is a routine strategy for screening potential drug candidates without the need for larger or longer studies designed to assess clinical end points. Understandably, surrogate end points also are attractive for use in clinical trials, especially if the surrogate occurs sooner, more often, or is easier to measure than a clinically relevant outcome. After results of phase 1 and 2 studies (often based on nonclinical end points), the phase 3 clinical trials necessary for drug approval ideally involve studying the drug’s effect on clinical end points. However, regulatory agencies will accept certain surrogate end points as substitutes for clinical end points in the regulatory process. Although the original rationale for this was to make promising new drugs available more quickly (particularly those used to treat life-threatening conditions such as human immunodeficiency virus/acquired immunodeficiency syndrome and certain forms of cancer), regulatory agencies now do this more commonly for medications used in diseases for which the need is less urgent.18 We argue that this is the wrong approach given the limited utility of surrogate end points, as noted previously and shown through the following examples.

Surrogate End Points or Clinical Correlates? Examples From Internal Medicine 

Class 1c Antiarrhythmics for Ventricular Premature Beats After Myocardial Infarction 

Perhaps the most widely quoted failure of nonclinical end points occurred when several antiarrhythmic agents were licensed for use post–myocardial infarction (MI) based on their ability to suppress ventricular premature beats (VPBs; a putative surrogate end point substituting for the clinical end point of sudden death19). When the Cardiac Arrhythmia Suppression Trial subsequently was performed, comparing the efficacy of 3 antiarrhythmic agents with placebo in patients with frequent VPBs after MI, the rate of sudden death was significantly higher in patients administered the antiarrhythmic agents.20 Although the frequency of VPBs after MI was known to correlate with poor clinical outcomes, suppression of VPBs turned out to be an invalid surrogate end point for prevention of sudden death. To highlight the magnitude of this mistake, an extra death resulted from every 21 patients treated over 10 months with the study drug. In addition, it is estimated that more Americans were killed as a result of class Ic antiarrhythmic agents used in this manner than were killed in the Vietnam war.21

Fluoride for Increasing Bone Density 

Another well-publicized failure of surrogate end points is the example of sodium fluoride treatment for osteoporosis in postmenopausal women. This agent, known to increase bone mass, was tested in an RCT of 202 postmenopausal women followed up for 4 years.16 Although bone mineral density in the lumbar spine increased by 35% (P < 0.001), new vertebral and nonvertebral fractures were more likely to occur in patients administered fluoride in comparison to those administered placebo. In the case of sodium fluoride, bone mineral density was not a valid surrogate end point for fracture reduction.

Hormone Replacement Therapy in Postmenopausal Women 

Hormone replacement therapy (HRT) had been advocated for postmenopausal women based on evidence of efficacy from small randomized trials examining the effect of HRT on “putative surrogate markers” (ie, lipoprotein levels) and also based on evidence from large observational trials.22, 23 However, 2 large randomized clinical trials using clinical end points24, 25 subsequently showed that HRT caused more harm than benefit. For example, the Women’s Health Initiative Study estimated absolute excess risks per 10,000 person-years attributable to estrogen plus progestin to be 7 more coronary heart disease events, 8 more strokes, 8 more pulmonary emboli, and 8 more invasive breast cancers, whereas estimated benefits were 6 fewer colorectal cancers and 5 fewer hip fractures per 10,000 person-years. Because of the frequent use of HRT during the last few decades, the failure of this surrogate end point led to a substantial number of potentially avoidable cardiovascular events in postmenopausal women.

These examples from other medical specialties highlight the potential dangers of relying exclusively on surrogate end points when assessing the effectiveness of new therapies. Nephrologists are particularly prone to errors resulting from the use of nonclinical end points because of the small number of randomized trials examining clinically important end points in patients with kidney disease.26 Definitive studies may never be performed unless past trends are reversed with respect to the use of surrogate end points in nephrology. Next, we highlight past problems and potential future issues in nephrology related to the use of surrogate end points.

Surrogate End Points or Clinical Correlates? Examples From Nephrology 

Changes in Glomerular Filtration Rate 

In the African American Study of Kidney Disease Trial,27 African Americans with hypertensive nephrosclerosis treated with a calcium channel blocker (amlodipine) experienced a slower mean decrease in estimated glomerular filtration rate (GFR) compared with those administered an angiotensin-converting enzyme (ACE) inhibitor (ramipril). Conversely, the ACE-inhibitor group had significantly lower rates of progression to the clinical end point of most interest (ESRD). The seeming paradox between the putative surrogate end point of GFR and the clinical end point resulted from the strong protective effect of ACE inhibitors in subjects at greatest risk for kidney failure and that amlodipine was found to increase GFR acutely because of a hemodynamic effect.27 Thus, relying on the GFR measure alone would have led to the erroneous conclusion that amlodipine was the superior agent for treatment of African Americans with hypertensive nephrosclerosis and proteinuria. In this example, GFR is a clinical correlate, rather than a true surrogate end point.

Reduction in Proteinuria 

In a position paper released by the National Kidney Foundation and the National Institute of Diabetes and Digestive and Kidney Diseases in 2003, the investigators noted that “accrued data show a strong relationship between level of proteinuria or albuminuria and progression of [chronic kidney disease] CKD to kidney failure and fatal cardiovascular disease events.”28 In a variety of glomerular conditions, level of proteinuria was shown to predict the occurrence of ESRD.29, 30, 31, 32 Furthermore, post hoc analyses from RCTs of patients with CKD administered ACE inhibitors showed that the relative risk reduction in patients with ESRD is predicted strongly by percentage of decrease in proteinuria from baseline.31, 33 In other words, patients for whom proteinuria decreases significantly are likely to experience the greatest benefit from therapy. This may explain the recent appearance of several RCTs testing the ability of certain interventions to decrease proteinuria, rather than prevent progression to ESRD.34, 35, 36 However, it is crucial to note that the association between proteinuria decrease and ESRD is independent of the study intervention, and the clinical benefit of strategies aimed at decreasing proteinuria per se is unknown.28 In the same National Kidney Foundation/National Institute of Diabetes and Digestive and Kidney Diseases jointly released position paper, the investigators correctly noted that present data do not enable assessment of whether decrease in proteinuria is a surrogate end point. They recommend performance of “new clinical trials to explore the potential validity of albuminuria as a surrogate marker of progression of CKD.”28

Abnormalities in Mineral Metabolism: Serum Phosphate and PTH Levels 

Returning to our illustrative example of sevelamer and cinacalcet in the setting of ESRD, we examine whether serum phosphate and PTH levels currently are established as valid surrogate end points. Given the noted context dependency of nonclinical end points, the examples of sevelamer and cinacalcet are considered separately. In our example, a literature search suggested that use of sevelamer is associated with slightly lower (0.4 mg/dL) serum calcium levels and slower rates of progression of coronary calcification scores in patients with ESRD,4 and a clinical trial presented at a recent meeting showed no significant differences in overall or cardiovascular mortality for patients administered sevelamer and calcium-based phosphate binders.37 Although clinical benefit was claimed in certain patient subgroups (ie, patients ≥ 65 years), these findings should be viewed as speculative given that the primary analysis was negative and nearly half the patients were lost to follow-up. In this case, it does not appear that coronary calcification is a valid surrogate end point for reduction in mortality associated with sevelamer. If subsequent studies confirm a clinical benefit in the elderly patient subgroup, this will serve to highlight the potential context dependency of surrogate end points (ie, acting as a surrogate end point in older, but not younger, patients), further showing the limited utility of surrogate end points.

As noted in the example, studies showed that cinacalcet decreases PTH levels in patients with ESRD with secondary hyperparathyroidism without precipitating hypercalcemia or hyperphosphatemia.7 Although a recent post hoc analysis of unvalidated outcomes from pooled clinical trials support the hypothesis that cinacalcet may decrease fracture and hospitalization rates,38 this remains speculative at present. Therefore, whether PTH level is a clinical correlate or a surrogate end point for cinacalcet must await results of adequately powered randomized trials that consider such clinical end points as mortality, cardiovascular morbidity, and bone fractures.

Implications of Surrogate Outcomes for Payers of Health Care 

In the setting of new opportunities for Medicare beneficiaries to receive discounts for medications and the disproportionate enrollment of patients with CKD in Medicare programs, the issues of medication coverage, costs, and opportunity costs are germane. Because no health care delivery system can afford all available beneficial therapies, strategic choices are required about which ones to support.39 From a societal perspective, resource use and product costs should be included in criteria for decisions to adopt novel therapies. In the setting of a fixed health care budget, payers of health services can support a new therapy only if others are not forgone that might impact on clinical outcomes to an equivalent extent, but at a reduced price, or that improve clinical outcomes to a greater extent at an equivalent price. Benefits associated with forgone health care programs or interventions constitute opportunity costs39 (Fig 3), often overlooked in the discussion of novel treatments, especially after such therapies have achieved regulatory approval.

Although regulatory agencies consider only the impact of a new drug on efficacy and safety, most funding bodies also consider the impact of a new therapy on costs (ie, the drug’s cost-effectiveness) in light of what treatment options currently are funded for such patients. This approach is a required component of drug funding in some countries (such as the United Kingdom, Australia, and Canada), but currently is not considered by the Medicare Coverage Advisory Committee in the United States. Although regulatory agencies assess the efficacy of a new therapy in isolation (ie, comparing only with placebo or standard care), funding bodies must assess health benefits and costs of the new drug under assessment with interventions currently funded, for both similar or unrelated patient groups and problems. The use of nonclinical end points, particularly ones not considered valid surrogate end points, is particularly challenging for decision makers and health care funding agencies because (without proof of clinically meaningful benefit) the opportunity cost associated with the new drug cannot be determined. When evaluating novel therapies, reliance on surrogates, rather than clinically relevant outcomes, could result in incorrect resource allocation. For example, if cinacalcet, sevelamer, or both improve markers of mineral metabolism without improving patient outcomes, widespread uptake of these medications would lead to large expenditures without meaningful benefit to patients. Given the high costs noted for both these agents and the paucity of data on clinical end points, it is not surprising that such agents currently are not reimbursed (except perhaps in restricted clinical situations) in Canada, the United Kingdom, or Australia.

Implications of Surrogate End Points for CPGs 

Although drugs may be registered on the basis of surrogate end points (or even possibly biomarkers) and without consideration of their costs, it would appear prudent to consider these issues during development of evidence-based CPGs. Because some CPGs may incorrectly endorse a clinical correlate as a clinical outcome, substantial rigor in literature interpretation is mandatory to avoid misapplication of clinical trial data. We believe it also is important to consider the issue of opportunity costs during the process of CPG development. It would seem prudent for CPG developers to refrain from recommending the use of new agents until they have been proved to improve clinically meaningful outcomes. This is particularly relevant because once a drug is in routine clinical use, it is unusual for physicians to change their prescribing practice in the absence of studies showing clear harm. In the case of sevelamer, given the expense and lack of information on clinical end points, it is surprising that recent CPGs recommend routine use of sevelamer in several common clinical situations.40

How Should Drugs Approved Without Clinical End Point Data Be Funded? 

Given these complexities, regulatory certification of a new drug does not and should not ensure funding through private or government-sponsored health care. Because financial pressures on health care systems are well known, policymakers must consider the strength of the clinical evidence (including the strength of the data in support of the surrogate end point), as well as cost implications. It is beyond the scope of this article to propose detailed solutions to resolve the competing demands of patients, physicians (who request funding for all new drugs that may provide potential clinical benefit), and payers of health care (who must maximize the health of all patients who receive care within their mandate). However, funding new drugs of uncertain clinical benefit clearly is not in the best interest of patients because this would remove companies’ incentives to perform high-quality studies that measure clinical end points. Without these data, it is unclear whether patients are receiving benefit (or even possibly harm) from the new drugs and whether new drugs provide good value for money.

The challenge is to develop policy that ensures that patients have access to therapies that provide clear benefit, but also gives incentive to companies to develop innovative and effective new drugs. One approach might involve temporary funding (or cost-sharing with industry) for new agents with regulatory approval shown to have a convincing impact on surrogate end points in a life-threatening condition. In this scenario, continued reimbursement would be conditional on subsequent proof that clinically meaningful outcomes also are improved. Although this suggestion may raise valid questions about feasibility, the Food and Drug Administration has used this approach for medications approved solely on the basis of surrogate end points, with the approval typically conditional upon the conduct of additional clinical studies verifying the clinical benefit of the medication. Therefore, the only change required would be to link ongoing funding to the findings of these trials.

Back to Article Outline

Conclusion 

To date, widespread change in nephrology practice typically has been driven by the findings of trials studying surrogate end points or even by observational data alone. Although this is historical fact, to be certain that we are providing the best possible care to patients with kidney failure, the nephrology community must demand and use data from large RCTs that consider clinically relevant end points. We suggest that before widespread uptake and unrestricted funding of any medication occurs, there is a need for randomized trial data showing that the new agent improves clinical end points. Proof of efficacy that is limited to improvements in surrogate end points is insufficient. Given the paucity of large RCTs evaluating interventions for patients with kidney disease, these considerations are highly relevant to nephrologists.

Back to Article Outline

References 

  1. Temple R . A regulatory authority’s opinion about surrogate endpoints . In:  Nimmo WS ,  Tucker GT editor. Clinical Measurement in Drug Evaluation . New York, NY: Wiley; 1995, pp 3-23;
  2. US Department of Health and Human Services: FDA: The Nation’s Premier Consumer Health Protection Agency, in US Food and Drug Administration, 2004. Available at: http://www.fda.gov/oc/opacom/brochure/Healthbro.pdf FDA 01-1316. Accessed November 1, 2006
  3. US Department of Health and Human Services: Guidance for Industry: Fast Track Drug Development Programs—Designation, Development, and Application Review, in, Food and Drug Administration, 2004. Available at: http://www.fda.gov/cder/guidance/index.htm. Accessed November 1, 2006
  4. Chertow GM , Burke SK , Raggi P . Sevelamer attenuates the progression of coronary and aortic calcification in hemodialysis patients . Kidney Int . 2002;62:245–252
  5. Huybrechts KF , Caro JJ , Wilson DA , O’Brien JA . Health and economic consequences of sevelamer use for hyperphosphatemia in patients on hemodialysis . Value Health . 2005;8:549–561
  6. Huybrechts KF , Caro JJ , London GM . Modeling the implications of changes in vascular calcification in patients on hemodialysis . Kidney Int . 2005;67:1532–1538
  7. Block GA , Martin KJ , de Francisco AL , et al.   Cinacalcet for secondary hyperparathyroidism in patients receiving hemodialysis . N Engl J Med . 2004;350:1516–1525
  8. Eknoyan G , Levin A , Levin NW . Bone metabolism and disease in chronic kidney disease . Am J Kidney Dis . 2003;42:1–201
  9. Fleming TR , DeMets DL . Surrogate end points in clinical trials (Are we being misled?) . Ann Intern Med . 1996;125:605–613
  10. Taylor JM , Wang Y . Surrogate markers and joint models for longitudinal and survival data . Control Clin Trials . 2002;23:626–634
  11. Lin DY , Fleming TR , De Gruttola V . Estimating the proportion of treatment effect explained by a surrogate marker . Stat Med . 1997;16:1515–1527
  12. Chen C , Wang H , Snapinn SM . Proportion of treatment effect (PTE) explained by a surrogate marker . Stat Med . 2003;22:3449–3459
  13. Prentice RL . Surrogate endpoints in clinical trials (Definition and operational criteria) . Stat Med . 1989;8:431–440
  14. Hughes MD . Evaluating surrogate endpoints . Control Clin Trials . 2002;23:703–707
  15. Burzykowski T , Molenberghs G , Buyse M . The Evaluation of Surrogate Endpoints . New York: Springer; 2005;
  16. Riggs BL , Hodgson SF , O’Fallon WM , et al.   Effect of fluoride treatment on the fracture rate in postmenopausal women with osteoporosis . N Engl J Med . 1990;322:802–809
  17. Biomarker Definitions Working Group: Biomarkers and surrogate endpoints: Preferred definitions and conceptual framework . Clin Pharmacol Ther . 2001;69:89–95
  18. Psaty BM , Weiss NS , Furberg CD , et al.   Surrogate end points, health outcomes, and the drug-approval process for the treatment of risk factors for cardiovascular disease . JAMA . 1999;282:786–790
  19. Pratt CM , Moye LA . The Cardiac Arrhythmia Suppression Trial (Background, interim results and implications) . Am J Cardiol . 1990;65:20B–29B
  20. The Cardiac Arrhythmia Suppression Trial (CAST) Investigators . Preliminary report (Effect of encainide and flecainide on mortality in a randomized trial of arrhythmia suppression after myocardial infarction) . N Engl J Med . 1989;321:406–412
  21. Moore T . Excess Mortality Estimates. Deadly Medicine: Why Tens of Thousands of Heart Patients Died in America’s Worst Drug Disaster . New York, NY: Simon & Schuster; 1995;
  22. Stampfer MJ , Colditz GA . Estrogen replacement therapy and coronary heart disease (A quantitative assessment of the epidemiologic evidence) . Prev Med . 1991;20:47–63
  23. Grady D , Rueben S , Pettiti D . Hormone therapy to prevent disease and prolong life in postmenopausal women . Ann Intern Med . 1992;117:1016–1037
  24. Hulley S , Grady D , Bush T , et al.  Heart and Estrogen/progestin Replacement Study (HERS) Research Group   Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women . JAMA . 1998;280:605–613
  25. Rossouw JE , Anderson GL , Prentice RL , et al.   Risks and benefits of estrogen plus progestin in healthy postmenopausal women (Principal results from the Women’s Health Initiative randomized controlled trial) . JAMA . 2002;288:321–333
  26. Strippoli GF , Craig JC , Schena FP . The number, quality, and coverage of randomized controlled trials in nephrology . J Am Soc Nephrol . 2004;15:411–419
  27. Wright JT , Bakris G , Greene T , et al.   Effect of blood pressure lowering and antihypertensive drug class on progression of hypertensive kidney disease (Results from the AASK Trial) . JAMA . 2002;288:2421–2431
  28. Eknoyan G , Hostetter T , Bakris GL , et al.   Proteinuria and other markers of chronic kidney disease (A position statement of the National Kidney Foundation (NKF) and the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK)) . Am J Kidney Dis . 2003;42:617–622
  29. Ruggenenti P , Perna A , Loriga G , et al.   Blood-pressure control for renoprotection in patients with non-diabetic chronic renal disease (REIN-2) (Multicentre, randomised controlled trial) . Lancet . 2005;365:939–946
  30. De Jong PE , de Zeeuw D . Renoprotective therapy (Is it blood pressure or albuminuria that matters?) . Lancet . 2005;365:913–914
  31. The GISEN Group (Gruppo Italiano di Studi Epidemiologici in Nefrologia) . Randomised placebo-controlled trial of effect of ramipril on decline in glomerular filtration rate and risk of terminal renal failure in proteinuric, non-diabetic nephropathy . Lancet . 1997;349:1857–1863
  32. Klahr S , Levey AS , Beck GJ , et al.  Modification of Diet in Renal Disease Study Group   The effects of dietary protein restriction and blood-pressure control on the progression of chronic renal disease . N Engl J Med . 1994;330:877–884
  33. Ravid M , Lang R , Rachmani R , Lishner M . Long-term renoprotective effect of angiotensin-converting enzyme inhibition in non-insulin-dependent diabetes mellitus. A 7-year follow-up study . Arch Intern Med . 1996;156:286–289
  34. Jacobsen P , Andersen S , Rossing K , Jensen BR , Parving HH . Dual blockade of the renin-angiotensin system versus maximal recommended dose of ACE inhibition in diabetic nephropathy . Kidney Int . 2003;63:1874–1880
  35. Rossing K , Schjoedt KJ , Jensen BR , Boomsma F , Parving HH . Enhanced renoprotective effects of ultrahigh doses of irbesartan in patients with type 2 diabetes and microalbuminuria . Kidney Int . 2005;68:1190–1198
  36. Jacobsen P , Andersen S , Jensen BR , Parving HH . Additive effect of ACE inhibition and angiotensin II receptor blockade in type I diabetic patients with diabetic nephropathy . J Am Soc Nephrol . 2003;14:992–999
  37. Suki WN , Zabeneh R , Cangiano J , et al.   The DCOR Trial (A prospective randomized trial assessing the impact on outcomes of sevelamer in dialysis patients) . J Am Soc Nephrol . 2005;16(suppl):S281A; (abstr)
  38. Cunningham J , Danese M , Olson K , Klassen P , Chertow GM . Effects of the calcimimetic cinacalcet HCl on cardiovascular disease, fracture, and health-related quality of life in secondary hyperparathyroidism . Kidney Int . 2005;68:1793–1800
  39. Manns B , Taub K , Donaldson C . Economic evaluation and end-stage renal disease (From basics to bedside) . Am J Kidney Dis . 2000;36:12–28
  40. National Kidney Foundation . K/DOQI Clinical Practice Guidelines for Bone Metabolism and Disease in Chronic Kidney Disease . Am J Kidney Dis . 2003;42(suppl 3):S1–S201

 Originally published online as doi:10.1053/j.ajkd.2006.03.044 on May 10, 2006.Support: None. Potential conflicts of interest: None.

PII: S0272-6386(06)00545-2

doi:10.1053/j.ajkd.2006.03.044

American Journal of Kidney Diseases
Volume 48, Issue 1 , Pages 159-166, July 2006

Article Tools

* Image Usage & Resolution