0
We're unable to sign you in at this time. Please try again in a few minutes.
Retry
We were able to sign you in, but your subscription(s) could not be found. Please try again in a few minutes.
Retry
There may be a problem with your account. Please contact the AMA Service Center to resolve this issue.
Contact the AMA Service Center:
Telephone: 1 (800) 262-2350 or 1 (312) 670-7827  *   Email: subscriptions@jamanetwork.com
Error Message ......
Original Investigation |

Serum Urea Nitrogen, Creatinine, and Estimators of Renal Function Mortality in Older Patients With Cardiovascular Disease FREE

Grace L. Smith, MD, MPH; Michael G. Shlipak, MD, MPH; Edward P. Havranek, MD; JoAnne M. Foody, MD; Frederick A. Masoudi, MD, MSPH; Saif S. Rathore, MPH; Harlan M. Krumholz, MS, MD
[+] Author Affiliations

Author Affiliations: Section of Cardiovascular Medicine, Department of Internal Medicine (Drs Smith, Foody, and Krumholz and Mr Rathore), Department of Epidemiology and Public Health (Dr Krumholz), and Robert Wood Johnson Clinical Scholars Program (Dr Krumholz), Yale University School of Medicine, and Center for Outcomes Research and Evaluation, Yale–New Haven Hospital (Drs Foody and Krumholz), New Haven, Conn; General Internal Medicine Section, Medical Service, San Francisco Veterans Affairs Medical Center, San Francisco, Calif (Dr Shlipak); Departments of Medicine, Epidemiology, and Biostatistics, University of California, San Francisco (Dr Shlipak); Division of Cardiology, Denver Health Medical Center, and Divisions of Cardiology and Geriatric Medicine, University of Colorado Health Sciences Center, Denver, Colo (Drs Havranek and Masoudi); and Colorado Foundation for Medical Care, Aurora (Dr Masoudi).


Arch Intern Med. 2006;166(10):1134-1142. doi:10.1001/archinte.166.10.1134.
Text Size: A A A
Published online

Background  Renal dysfunction predicts increased mortality in cardiovascular patients, but the best renal estimator for quantifying risks is uncertain. We compared admission serum urea nitrogen (SUN) level, creatinine level, Modification of Diet in Renal Disease (MDRD) rate, and Mayo estimated glomerular filtration rate (eGFR) for predicting mortality.

Methods  In a retrospective cohort of Medicare patients (aged ≥65 years) hospitalized for myocardial infarction (n = 44 437) and heart failure (n = 56 652), renal estimators were compared for linearity with 1-year mortality risk, magnitude of risk, and relative importance for predicting risk (percentage variance explained) in proportional hazards models.

Results  The SUN level, creatinine level, and Mayo eGFR had linear associations with mortality. These measures predicted steadily increased risk in patients who experienced a myocardial infarction with a SUN level greater than 17 mg/dL (>6.1 mmol/L), a creatinine level greater than 1.0 mg/dL (>88.4 μmol/L), and a Mayo eGFR of less than 100 mL/min per 1.73 m2; and in patients who experienced heart failure with a SUN level greater than 16 mg/dL (>5.7 mmol/L), a creatinine level greater than 1.1 mg/dL (>97.2 μmol/L), and a Mayo eGFR of 90 mL/min per 1.73 m2 or less. In contrast, the MDRD eGFR had a J-shaped association and failed to identify increased risks in 50.0% of patients who experienced a myocardial infarction (with an MDRD eGFR >55 mL/min per 1.73 m2) and 60.0% of patients who experienced heart failure (with an MDRD eGFR >44 mL/min per 1.73 m2). The SUN level and Mayo eGFR had the greatest magnitude of risks. In myocardial infarction and heart failure patients, adjusted mortality increased by 3% and 7%, respectively, per 5-U increase in SUN, and by 3% and 9%, respectively, per 10-U decrease in Mayo eGFR (P<.001), based on models including both renal measures. Of all the measures, SUN had the greatest magnitude of relative importance for predicting mortality.

Conclusions  In older cardiovascular patients, SUN- and creatinine-based measures were powerful predictors of postdischarge mortality. Only MDRD eGFR was less adequate in quantifying risks for patients with mild impairment. Novel estimators, such as the Mayo eGFR, may play an important role in outcomes' prognostication for these patients.

Figures in this Article

Renal dysfunction is associated with increased mortality in patients with myocardial infarction (MI) and heart failure (HF),110 but the best measure for predicting mortality risk is uncertain. Studies of MI and HF patients have used various measures of renal function to predict mortality risk, including directly measured serum urea nitrogen (SUN) and creatinine levels and the estimated glomerular filtration rate (eGFR), calculated by the Modification of Diet in Renal Disease (MDRD) equation, which incorporates creatinine level with age, sex, and race.11,12 Of these commonly available measures, many national guidelines and experts consider MDRD eGFR to be the most valid measure of renal function and, thus, the best measure for assessing prognosis in cardiovascular patients.1316 However, to our knowledge, no empirical comparison of these measures has actually been conducted in a representative sample of older subjects, who compose most cardiovascular patients.

The nationally representative National Heart Care Project cohort of Medicare beneficiaries hospitalized for MI or HF provides an ideal opportunity to empirically compare measures of renal function. Accordingly, we evaluated SUN level, creatinine level, and MDRD eGFR, and the recently proposed Mayo eGFR,17 for predicting 1-year mortality, based on strength and magnitude of independent adjusted association with mortality and relative importance for explaining mortality.

NATIONAL HEART CARE PROJECT

The National Heart Care Project is a Centers for Medicare and Medicaid Services quality improvement initiative for patients hospitalized between March 1, 1998, and April 30, 1999, and July 1, 2000, and June 30, 2001, with a principal discharge diagnosis of MI (International Classification of Diseases, Ninth Revision, Clinical Modification codes 410.xx, excluding code 410.x2) or HF (codes 402.01, 402.11, 402.91, 404.01, 404.91, or 428). Up to 800 HF and 750 MI cases were randomly sampled by state in each period after sorting by age, race, sex, and hospital. Data validity and precision were ensured through the use of trained reviewers at central abstraction sites, specific abstraction software incorporating logic checks, and random record reabstraction. Hospital and physician data were derived from the American Hospital Association annual survey and the American Medical Association Physician Masterfile. Detailed methods are published elsewhere.1,18

PATIENT COHORTS

Excluded from the initial sample of 71 120 MI patients were those younger than 65 years (n = 6042), transfers from other hospitals because of a missing baseline renal function measure (n = 11 878), those with no clinical confirmation of MI (based on creatine kinase–MB or troponin level, symptom, and electrocardiographic criteria) (n = 7902), or those with a terminal illness (n = 133).

Excluded from the initial sample of 78 882 HF patients were second visits of patients in the sample appearing more than once (n = 3732); those younger than 65 years (n = 6558); transfers from other hospitals (n = 2419); those with no evidence of HF on admission by clinical symptoms or chest x-ray film (n = 5003); patients with aortic stenosis (n = 5493) and mitral stenosis (n = 243), to exclude valvular HF; and patients undergoing long-term renal dialysis (n = 549). Patients may have had 1 or more reasons for exclusion.

Patients with a missing admission creatinine level (MI group, n = 1089; and HF group, n = 1746); a missing SUN level (MI group, n = 1244; and HF group, n = 3362); missing, nonblack, or nonwhite race (effectively missing MDRD estimate) (MI group, n = 5401; and HF group, n = 3216); and unknown date of death (MI group, n = 303; and HF group, n = 969) were excluded from analysis, for a total sample of 44 437 in the MI cohort and 56 652 in the HF cohort.

RENAL FUNCTION

Measured renal function estimators included first-admission SUN and creatinine levels (measured in milligrams per deciliter) (based on tests within the first 6 hours of admission). Calculated GFR estimates included the simplified MDRD prediction equation12: 186 × (Serum Creatinine Level−1.154) × (Age−0.203) × 1.212 (If Black) × 0.742 (If Female), measured in milliliters per minute per 1.73 m2, abbreviated as MDRD eGFR in this article. The recently proposed Mayo Healthy-Chronic Kidney Disease prediction equation17 was also included: exp {[(1.911 + 5.249)/Serum Creatinine] − (2.114/Serum Creatinine2) − (0.00686 × Age) − [0.205 (If Female)]}, abbreviated as Mayo eGFR.

OUTCOMES AND COVARIATES

All-cause mortality was assessed using the Medicare enrollment database and the Medicare part A database. Up to 1-year follow-up was calculated from admission date.

Potential confounders, selected based on prior studies and clinical relevance, included age, race, and sex; history of HF, MI, hypertension, angina, coronary artery bypass graft, percutaneous coronary intervention, diabetes mellitus, stroke, smoking, chronic obstructive pulmonary disease, and dementia; mobility; presenting peripheral edema; heart rate, respiratory rate, and systolic blood pressure; left ventricular systolic function (normal, mild, moderate, severe, and not documented); admission serum sodium, potassium, glucose, hematocrit, and albumin levels; preadmission use of aspirin, angiotensin-converting enzyme inhibitors, angiotensin II receptor blockers, β-blockers, diuretics, digoxin, antiarrhythmics, and calcium channel blockers; prearrival setting; hospital setting (urban vs rural); and treating physician specialty. Covariates lacking linear relationships with outcomes were coded categorically as dummy variables in multivariable analyses, including a dummy variable for missing values. All variables had less than 5% missing values, except for “not documented” left ventricular systolic function.

STATISTICAL ANALYSIS
Linearity of Association

To compare renal function measures, we graphically compared unadjusted 1-year mortality risks by the 5th to 95th percentile values for SUN level, creatinine level, MDRD eGFR, and Mayo eGFR. Correlations between renal function measures were determined using the Pearson product-moment correlation coefficient.

Strength and Magnitude of Association

Second, we compared mortality risks associated with each renal function measure in separate proportional hazards models adjusted for all other covariates, with results shown for deciles of renal function for ease of presentation. Complementary log-log plots confirmed proportionality assumptions. Subsidiary models accounting for weighting by state and clustering by hospital and stratifying by age (dichotomized at the age of 75 years), sex, and race were conducted to verify the validity of models.

Relative Importance of Renal Function Measures

Third, we evaluated the relative prognostic importance of SUN level, creatinine level, MDRD eGFR, and Mayo eGFR compared with each other and with other predictors (such as systolic blood pressure, left ventricular systolic function, and age) in the multivariable models, based on the contribution of each renal function predictor (SUN level, creatinine level, MDRD eGFR, or Mayo eGFR) and the individual contribution of other covariates, to total explained variance in 1-year mortality risks. Type 3 Wald χ2 values were used to calculate marginal improvement in R2 values attributed to each predictor, and percentage explained variance was calculated based on the R2 of the full model containing all covariates as the denominator.19

Analyses were conducted using SAS statistical software, version 9.1 (SAS Institute Inc, Cary, NC), and Stata, version 7.0 (Stata Corp, College Station, Tex), and all statistical tests assumed a 2-tailed α = .05. Analyses were conducted by one of us (G.L.S.). Use of the National Heart Care Project database was approved by the Yale University School of Medicine Human Investigation Committee.

MI AND HF COHORTS

In 44 437 MI patients, the mean ± SD age was 78 ± 8 years, 36.8% had a prior MI, 32.3% had diabetes mellitus, and 70.1% had hypertension. In 56 652 HF patients, the mean ± SD age was 79 ± 8 years, 71.6% had a prior exacerbation, 39.8% had diabetes mellitus, and 63.5% had hypertension (Table 1). In both cohorts, patients presented with a spectrum of renal function, and all measures of renal function were significantly, although not perfectly, correlated. Pairwise comparisons among creatinine level, MDRD eGFR, and Mayo eGFR had correlations ranging from  = 0.69 to r = 0.83 (P<.001), and comparisons between SUN level and the other measures had correlations ranging from r = 0.59 to r = 0.73 (P<.001).

MORTALITY RISKS

By 1 year after discharge, 33.8% of MI patients and 37.7% of HF patients died. Worse renal function was associated with progressively higher mortality risks. The SUN level, creatinine level, and Mayo eGFR had approximately linear associations with mortality risks (Figure 1). Furthermore, in multivariable models, SUN level, creatinine level, and Mayo eGFR were each significant (P<.001) independent predictors of mortality, even in normal to near-normal ranges. Progressively and significantly increased mortality risk was associated with a SUN level greater than 17 mg/dL (>6.1 mmol/L), a creatinine level greater than 1.0 mg/dL (>88.4 μmol/L), and a Mayo eGFR of less than 100 mL/min per 1.73 m2 in MI patients; and with a SUN level greater than 16 mg/dL (>5.7 mmol/L), a creatinine level greater than 1.1 mg/dL (>97.2 μmol/L), and a Mayo eGFR of 90 mL/min per 1.73 m2 or less in HF patients (Table 2 and Table 3, respectively).

Place holder to copy figure label and caption
Figure 1.

Mortality curves for 5th to 95th percentile values of serum urea nitrogen (SUN) level (A and B), creatinine level (C and D), Modification of Diet in Renal Disease (MDRD) estimated glomerular filtration rate (eGFR) (E and F), and Mayo eGFR in patients who experienced a myocardial infarction (A, C, E, and G) and heart failure (B, D, F, and H). To convert SUN to millimoles per liter, multiply by 0.357; and to convert creatinine to micromoles per liter, multiply by 88.4.

Graphic Jump Location
Table Graphic Jump LocationTable 2. Data for SUN, Creatinine, and eGFR: Associated Mortality Risks in Patients With MI*
Table Graphic Jump LocationTable 3. Data for SUN, Creatinine, and eGFR: Associated Mortality Risks in Patients With HF*

In contrast, MDRD eGFR had a J-shaped association with mortality and failed to identify increased mortality risks in those patients with a normal to near-normal eGFR (>60 mL/min per 1.73 m2). Specifically in multivariable models, the MDRD eGFR provided no additional prognostic information for the 50.0% of MI patients with an MDRD eGFR greater than 55 mL/min per 1.73 m2 or the 60.0% of HF patients with an MDRD eGFR greater than 44 mL/min per 1.73 m2 (Figure 1 and Table 2 and Table 3, respectively).

Of all 4 measures, SUN level and Mayo eGFR had the greatest magnitude of adjusted mortality risks (Table 2 and Table 3). In multivariable models that included SUN level and Mayo eGFR as continuous variables, both remained independent significant (P<.001) predictors of mortality. For example, after adjusting for Mayo eGFR, the incremental 1-year mortality risk in MI patients increased by 3% per 5-U increase in SUN level (hazard ratio, 1.03; 95% confidence interval, 1.02-1.04; P<.001); and in HF patients, it increased by 7% per 5-U increase in SUN level (hazard ratio, 1.07; 95% confidence interval, 1.07-1.08; P<.001). This translates into a 12% relative increase in mortality risk for a 20-U increase in SUN level for MI patients and a 33% relative increase for HF patients (Table 4). Weighting and clustering by state and hospital did not alter estimates.

Table Graphic Jump LocationTable 4. Incremental Adjusted Mortality Risks Associated With Elevated SUN and Decreased Mayo eGFR
RELATIVE IMPORTANCE OF RENAL FUNCTION MEASURES

Of all 4 measures, the SUN level explained the greatest variance in mortality risks, even after adjusting for the effect of creatinine-based measures (creatinine level, MDRD eGFR, or Mayo eGFR). Particularly in HF patients, the SUN level rivaled systolic blood pressure, left ventricular systolic function, and age in relative prognostic importance and was the most important of any predictor of mortality (Figure 2). A subsidiary analysis using the SUN-creatinine ratio found that it was no better than SUN level for explaining variance in mortality risks. The relative comparison of renal function measures was similar in subsidiary analyses stratified by age, sex, and race.

Place holder to copy figure label and caption
Figure 2.

Relative prognostic importance of serum urea nitrogen (SUN) level, creatinine level, Modification of Diet in Renal Disease estimated glomerular filtration rate (MDRD eGFR), Mayo eGFR and other risk factors as indicated by percentage of explained variance of mortality risks in patients who experienced a myocardial infarction (MI) (A) and heart failure (HF) (B). The asterisk indicates that there was an improvement in explained variance (R2) compared with a model without the predictor of interest, relative to the explained variance for the full model with all the covariates (actual value for the MI group, R2 = 0.25; and for the HF group, R2 = 0.20). These models were adjusted for age, sex, race, comorbidities, medications, and hospital characteristics. All variables were significant predictors of mortality at P<.001, except for prior MI (P<.05). The dagger indicates this was an alternative model using Mayo eGFR instead of MDRD eGFR. ACEI indicates angiotensin-converting enzyme inhibitor; and LVSF, left ventricular systolic function.

Graphic Jump Location

In older cardiovascular patients hospitalized for MI and HF, SUN level and creatinine-based measures each had independent value for predicting 1-year postdischarge mortality. Furthermore, particularly in HF patients, SUN level rivaled the prognostic importance of traditional factors used to risk stratify cardiovascular patients, such as systolic blood pressure, ejection fraction, and age. Although SUN level and most creatinine-based estimators provided significant, incremental, predictive information across their entire ranges, surprisingly, the MDRD estimate performed poorly for identifying increased mortality risks, particularly among patients with an eGFR greater than 60 mL/min per 1.73 m2.

ROLE OF eGFR IN CARDIOVASCULAR PATIENTS

With increasing recognition of renal dysfunction as a potential cardiovascular disease equivalent, guidelines from the American Heart Association,14 the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure,20 the National Kidney Foundation,16 and the National Kidney Disease Education Program13 have recommended routine assessment of renal function in patients with cardiovascular risk factors or recognized cardiovascular disease using creatinine-based prediction equations,14 particularly the MDRD calculation.13,16 Given these recommendations, some laboratories have begun routine reporting of MDRD eGFR and numerous studies2,3,8,21 of HF and MI patients have attempted to risk stratify patients using the MDRD eGFR.

Recently, Anavekar et al2 reported a 10% increase in risk of mortality and nonfatal cardiovascular events for every 10-U decrease of MDRD eGFR below 81 mL/min per 1.73 m2 in patients with MI and HF or left ventricular systolic dysfunction, prompting assertions that the MDRD equation is the best of commonly available estimates of renal function in cardiovascular patients, and a valid and powerful predictor of outcomes in this group.15 The MDRD calculation provides a more valid estimate of true GFR than direct serum creatinine measurement and the traditional Cockcroft-Gault creatinine clearance estimate,22 and furthermore avoids the difficult to obtain ideal body weight12 required for Cockcroft-Gault creatinine clearance23 (excluded from our analysis also because of lack of weight data). However, the validity and generalizability of this equation has also been questioned,17 particularly because the MDRD equation does not effectively discriminate among persons with an eGFR of greater than 60 mL/min per 1.73 m2.24

The role of the MDRD equation in mortality risk prediction has particularly been debated,17,21,2529 especially for patients with mild renal dysfunction,28,30,31 because this equation was derived in a clinical trial population that was relatively young and also had relatively severe renal impairment. Limitations of the MDRD eGFR are especially concerning when applying the equation to risk stratification of cardiovascular patients, given that many of these patients have only mild renal impairment yet still have appreciable increased mortality risk. The J-shaped association with mortality found in our study may be because of overestimation or misclassification of true GFR in this range, particularly for elderly patients, in whom lower creatinine values could indicate unmeasured frailty or comorbidity, reflected by lower muscle mass.

Interestingly, although originally derived from a relatively limited patient population, in our study cohort, the Mayo eGFR equation performed quite well in identifying an excess mortality risk for patients with an eGFR of greater than 60 mL/min per 1.73 m2, suggesting that the Mayo eGFR estimate warrants further consideration as a clinical tool.3234 However, further validation of this equation is also necessary because this formula was based on a relatively young and racially homogeneous population.

ROLE OF SUN LEVEL

Despite the difficulties inherent in characterizing GFR in cardiovascular patients, studies have focused on creatinine-based measures for risk stratification, while SUN level has been generally underappreciated as an important prognostic renal function measure. In fact, most clinical trials of cardiovascular patients rely on creatinine level alone for inclusion, exclusion, and stratification criteria. Because SUN level can simultaneously reflect alterations in GFR and global volume, it is the least precise estimator of GFR. Nevertheless, in our study, it was at least as important a prognostic indicator for mortality as eGFR, especially in HF patients, likely because it is an independent marker of HF severity. Moreover, SUN level is particularly important in this elderly population because creatinine may weaken its ability to reflect changes in GFR as muscle mass decreases.

Prior studies35,36 of smaller cohorts of HF patients have provided conflicting results regarding the significance of SUN level for predicting long-term mortality. Fonarow et al37 recently showed that an extreme elevation in SUN level (≥43 mg/dL [≥15.4 mmol/L]) was the most important predictor of in-hospital mortality in HF patients with acute decompensation. This study, however, did not address whether less severe elevations in SUN level could also be meaningful, especially after considering the effect of eGFR, and did not explore whether SUN level was useful for risk stratification in long-term follow-up. In our study, no single cut point for SUN level to predict mortality risk existed, because the incremental risks across the whole spectrum were informative, alongside the incremental risks across the whole spectrum of eGFR.

In MI patients, many risk scores have not included renal function.3841 The PREDICT42 (Predicting Risk of Death in Cardiac Disease Tool) and Cooperative Cardiovascular Project43 risk scores for long-term outcomes consider severe elevations in SUN or creatinine level. Yet, similar to most studies of HF patients, they fail to take advantage of the rich prognostic information provided by a continuous spectrum of renal function, particularly in the presumed “normal” range.44 Therefore, our novel comparison of renal function measures may help to prompt reevaluation of current risk scores in MI and HF patients, first to account for the full range of renal impairment, second to consider the use of novel estimators of GFR in patients with only mild dysfunction, and third to include elevations in SUN level as a unique predictor distinct from GFR. Because SUN level may reflect multiple facets of cardiorenal pathophysiological features, particularly in the short-term setting of hospitalization, it may maintain a significant role in risk stratifying cardiovascular patients even as promising assays for estimating GFR, such as cystatin C, potentially supersede existing creatinine-based measures in outpatients.45

LIMITATIONS

Our cohorts included only older Medicare beneficiaries with baseline renal estimators measured in the hospital and, thus, may not be generalizable to younger MI and HF patients with renal function assessed in a stable outpatient context. The contribution of acute vs chronic and intrinsic vs prerenal dysfunction could not be distinguished in our study; however, even simple single measures of renal function remained surprisingly linear and robust in predicting outcomes. Our study did not have a gold standard measurement of GFR and was based on creatinine measurements from numerous hospitals (thus, nonstandardized). However, our study demonstrates the predictive ability that can be expected of these renal function estimates in hospitals across the country based on available methods of measurement. Therefore, it is not intended to identify the most precise estimator of GFR, only to evaluate renal function measures as predictors of mortality risks. However, the role of empirical validation of renal function measures used in clinical stratification for predicting outcomes can serve as an important complement to physiologically validating estimates with actual gold standards.

In conclusion, the empirical comparison of renal function measures is important for informing current practice for risk stratifying cardiovascular patients, because up to half of older MI and HF patients have some degree of renal dysfunction. The entire ranges of SUN level and GFR are powerful predictors of 1-year mortality following hospitalization for MI or HF. Optimal cardiovascular disease risk prediction should consider including SUN level and eGFR, with novel estimators of GFR, such as the Mayo eGFR equation, playing a new and important role.

Correspondence: Harlan M. Krumholz, MS, MD, Robert Wood Johnson Clinical Scholars Program, Yale University School of Medicine, 333 Cedar St, Sterling Hall of Medicine, I-Wing, Suite 456, New Haven, CT 06520 (harlan.krumholz@yale.edu).

Accepted for Publication: January 13, 2006.

Financial Disclosure: None.

Funding/Support: This study was supported by Medical Scientist Training grant GM07205 from the National Institute of General Medical Sciences, National Institutes of Health (Dr Smith and Mr Rathore); the Paul Beeson Scholars Program of the American Federation for Aging Research (Dr Shlipak); the Generalist Faculty Scholars Program of the Robert Wood Johnson Foundation (Dr Shlipak); Research Career Awards K08-AG20623-01 (Dr Foody) and K08-AG01011 (Dr Masoudi) from the National Institute on Aging, National Institutes of Health; and a fellowship in geriatrics from the National Institute on Aging/John A. Hartford Foundation (Dr Foody).

Role of the Sponsor: The funding bodies had no role in data extraction and analyses, in the writing of the manuscript, or in the decision to submit the manuscript for publication.

Disclaimer: The analyses on which this publication is based were performed under contract 500-02-CO01, entitled “Utilization and Quality Control Peer Review Organization for the State (Commonwealth) of Colorado,” sponsored by the Centers for Medicare and Medicaid Services (formerly the Health Care Financing Administration), Department of Health and Human Services. The content of the publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the US government. The authors assume full responsibility for the accuracy and completeness of the ideas presented. This article is a direct result of the Health Care Quality Improvement Program initiated by the Centers for Medicare and Medicaid Services, which has encouraged identification of quality improvement projects derived from analysis of patterns of care and, therefore, required no special funding on the part of this contractor. Ideas and contributions to the authors concerning experience in engaging with issues presented are welcomed.

Smith  GLShlipak  MGHavranek  EP  et al.  Race and renal impairment in heart failure: mortality in blacks versus whites. Circulation 2005;1111270- 1277
Link to Article
Anavekar  NSMcMurray  JJVelazquez  EJ  et al.  Relation between renal dysfunction and cardiovascular outcomes after myocardial infarction. N Engl J Med 2004;3511285- 1295
PubMed Link to Article
Shlipak  MGSmith  GLRathore  SSMassie  BMKrumholz  HM Renal function, digoxin therapy, and heart failure outcomes: evidence from the Digoxin Intervention Group trial. J Am Soc Nephrol 2004;152195- 2203
PubMed Link to Article
Smith  GLVaccarino  VKosiborod  M  et al.  Worsening renal function: what is a clinically meaningful change in creatinine during hospitalization with heart failure? J Card Fail 2003;913- 25
PubMed Link to Article
Dries  DLExner  DVDomanski  MJGreenberg  BStevenson  LW The prognostic implications of renal insufficiency in asymptomatic and symptomatic patients with left ventricular systolic dysfunction. J Am Coll Cardiol 2000;35681- 689
PubMed Link to Article
Hillege  HLGirbes  ARde Kam  PJ  et al.  Renal function, neurohormonal activation, and survival in patients with chronic heart failure. Circulation 2000;102203- 210
PubMed Link to Article
McAlister  FAEzekowitz  JTonelli  MArmstrong  PW Renal insufficiency and heart failure: prognostic and therapeutic implications from a prospective cohort study. Circulation 2004;1091004- 1009
PubMed Link to Article
Al-Ahmad  ARand  WMManjunath  G  et al.  Reduced kidney function and anemia as risk factors for mortality in patients with left ventricular dysfunction. J Am Coll Cardiol 2001;38955- 962
PubMed Link to Article
Shlipak  MGHeidenreich  PANoguchi  HChertow  GMBrowner  WSMcClellan  MB Association of renal insufficiency with treatment and outcomes after myocardial infarction in elderly patients. Ann Intern Med 2002;137555- 562
PubMed Link to Article
Wright  RSReeder  GSHerzog  CA  et al.  Acute myocardial infarction and renal dysfunction: a high-risk combination. Ann Intern Med 2002;137563- 570
PubMed Link to Article
Levey  ASBosch  JPLewis  JBGreene  TRogers  NRoth  DModification of Diet in Renal Disease Study Group, A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Ann Intern Med 1999;130461- 470
PubMed Link to Article
Levey  ASGreene  TKusek  JWBeck  GJ A simplified equation to predict glomerular filtration rate from serum creatinine [abstract]. J Am Soc Nephrol 2000;11155A
 National Kidney Disease Education Program (NKDEP). http://www.nkdep.nih.gov. Accessed March 3, 2005
Sarnak  MJLevey  ASSchoolwerth  AC  et al.  Kidney disease as a risk factor for development of cardiovascular disease: a statement from the American Heart Association Councils on Kidney in Cardiovascular Disease, High Blood Pressure Research, Clinical Cardiology, and Epidemiology and Prevention. Circulation 2003;1082154- 2169
PubMed Link to Article
Hostetter  TH Chronic kidney disease predicts cardiovascular disease. N Engl J Med 2004;3511344- 1346
PubMed Link to Article
Levey  ASCoresh  JBalk  E  et al.  National Kidney Foundation practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Ann Intern Med 2003;139137- 147
PubMed Link to Article
Rule  ADLarson  TSBergstralh  EJSlezak  JMJacobsen  SJCosio  FG Using serum creatinine to estimate glomerular filtration rate: accuracy in good health and in chronic kidney disease. Ann Intern Med 2004;141929- 937
PubMed Link to Article
Foody  JMFerdinand  FDGalusha  D  et al.  Patterns of secondary prevention in older patients undergoing coronary artery bypass grafting during hospitalization for acute myocardial infarction. Circulation 2003;108 ((suppl 1)) II24- II28
PubMed Link to Article
Schemper  M The relative importance of prognostic factors in studies of survival. Stat Med 1993;122377- 2382
PubMed Link to Article
Chobanian  AVBakris  GLBlack  HR  et al.  The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 report. JAMA 2003;2892560- 2572
PubMed Link to Article
Go  ASChertow  GMFan  DMcCulloch  CEHsu  CY Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. N Engl J Med 2004;3511296- 1305
PubMed Link to Article
Shemesh  OGolbetz  HKriss  JPMyers  BD Limitations of creatinine as a filtration marker in glomerulopathic patients. Kidney Int 1985;28830- 838
PubMed Link to Article
Cockcroft  DWGault  MH Prediction of creatinine clearance from serum creatinine. Nephron 1976;1631- 41
PubMed Link to Article
Stevens  LALevey  AS Clinical implications of estimating equations for glomerular filtration rate. Ann Intern Med 2004;141959- 961
PubMed Link to Article
Froissart  MRossert  JJacquot  CPaillard  MHouillier  P Predictive performance of the Modification of Diet in Renal Disease and Cockcroft-Gault equations for estimating renal function. J Am Soc Nephrol 2005;16763- 773
PubMed Link to Article
Smilde  TDHillege  HLNavis  GVoors  AAvan Veldhuisen  DJ Formulas estimating renal function give imprecise estimates in heart failure [abstract]. J Am Coll Cardiol 2005;45172A
Link to Article
Risch  LSagmeister  MHuber  A Chronic renal disease and cardiovascular risk [letter]. N Engl J Med 2005;352199- 200
PubMed Link to Article
Cheng  H Chronic renal disease and cardiovascular risk [letter]. N Engl J Med 2005;352199- 200
PubMed Link to Article
Lamb  EJWebb  MCSimpson  DECoakley  AJNewman  DJO’Riordan  SE Estimation of glomerular filtration rate in older patients with chronic renal insufficiency: is the Modification of Diet in Renal Disease formula an improvement? J Am Geriatr Soc 2003;511012- 1017
PubMed Link to Article
Bostom  AGKronenberg  FRitz  E Predictive performance of renal function equations for patients with chronic kidney disease and normal serum creatinine levels. J Am Soc Nephrol 2002;132140- 2144
PubMed Link to Article
Corsonello  APedone  CIncalzi  RAGemelli  PA Methods for estimating glomerular filtration rate. JAMA 2004;2912819- 2820
PubMed
Maaravi  YBursztyn  MStessman  J The new Mayo Clinic equation for estimating glomerular filtration rate [letter]. Ann Intern Med 2005;142680- 681
PubMed Link to Article
Delanaye  PKrzesinski  JM The new Mayo Clinic equation for estimating glomerular filtration rate [letter]. Ann Intern Med 2005;142679- 680
PubMed Link to Article
Froissart  MCRossert  JHouillier  P The new Mayo Clinic equation for estimating glomerular filtration rate [letter]. Ann Intern Med 2005;142679
PubMed Link to Article
Felker  GMLeimberger  JDCaliff  RM  et al.  Risk stratification after hospitalization for decompensated heart failure. J Card Fail 2004;10460- 466
PubMed Link to Article
Lee  DSAustin  PCRouleau  JLLiu  PPNaimark  DTu  JV Predicting mortality among patients hospitalized for heart failure: derivation and validation of a clinical model. JAMA 2003;2902581- 2587
PubMed Link to Article
Fonarow  GCAdams  KF  JrAbraham  WTYancy  CWBoscardin  WJ Risk stratification for in-hospital mortality in acutely decompensated heart failure: classification and regression tree analysis. JAMA 2005;293572- 580
PubMed Link to Article
Califf  RMPieper  KSLee  KL  et al.  Prediction of 1-year survival after thrombolysis for acute myocardial infarction in the Global Utilization of Streptokinase and TPA for Occluded Coronary Arteries trial. Circulation 2000;1012231- 2238
PubMed Link to Article
Boersma  EPieper  KSSteyerberg  EW  et al. PURSUIT Investigators, Predictors of outcome in patients with acute coronary syndromes without persistent ST-segment elevation: results from an international trial of 9461 patients. Circulation 2000;1012557- 2567
PubMed Link to Article
Antman  EMCohen  MBernink  PJ  et al.  The TIMI risk score for unstable angina/non-ST elevation MI: a method for prognostication and therapeutic decision making. JAMA 2000;284835- 842
PubMed Link to Article
Morrow  DAAntman  EMCharlesworth  A  et al.  TIMI risk score for ST-elevation myocardial infarction: a convenient, bedside, clinical score for risk assessment at presentation: an intravenous nPA for treatment of infarcting myocardium early II trial substudy. Circulation 2000;1022031- 2037
PubMed Link to Article
Jacobs  DR  JrKroenke  CCrow  R  et al.  PREDICT: a simple risk score for clinical severity and long-term prognosis after hospitalization for acute myocardial infarction or unstable angina: the Minnesota heart survey. Circulation 1999;100599- 607
PubMed Link to Article
Krumholz  HMChen  JChen  YTWang  YRadford  MJ Predicting one-year mortality among elderly survivors of hospitalization for an acute myocardial infarction: results from the Cooperative Cardiovascular Project. J Am Coll Cardiol 2001;38453- 459
PubMed Link to Article
Kirtane  AJLeder  DMWaikar  SS  et al.  Serum blood urea nitrogen as an independent marker of subsequent mortality among patients with acute coronary syndromes and normal to mildly reduced glomerular filtration rates. J Am Coll Cardiol 2005;451781- 1786
PubMed Link to Article
Shlipak  MGSarnak  MJKatz  R  et al.  Cystatin-C and risk for mortality and cardiovascular disease in elderly adults. N Engl J Med 2005;3522049- 2060
PubMed Link to Article

Figures

Place holder to copy figure label and caption
Figure 1.

Mortality curves for 5th to 95th percentile values of serum urea nitrogen (SUN) level (A and B), creatinine level (C and D), Modification of Diet in Renal Disease (MDRD) estimated glomerular filtration rate (eGFR) (E and F), and Mayo eGFR in patients who experienced a myocardial infarction (A, C, E, and G) and heart failure (B, D, F, and H). To convert SUN to millimoles per liter, multiply by 0.357; and to convert creatinine to micromoles per liter, multiply by 88.4.

Graphic Jump Location
Place holder to copy figure label and caption
Figure 2.

Relative prognostic importance of serum urea nitrogen (SUN) level, creatinine level, Modification of Diet in Renal Disease estimated glomerular filtration rate (MDRD eGFR), Mayo eGFR and other risk factors as indicated by percentage of explained variance of mortality risks in patients who experienced a myocardial infarction (MI) (A) and heart failure (HF) (B). The asterisk indicates that there was an improvement in explained variance (R2) compared with a model without the predictor of interest, relative to the explained variance for the full model with all the covariates (actual value for the MI group, R2 = 0.25; and for the HF group, R2 = 0.20). These models were adjusted for age, sex, race, comorbidities, medications, and hospital characteristics. All variables were significant predictors of mortality at P<.001, except for prior MI (P<.05). The dagger indicates this was an alternative model using Mayo eGFR instead of MDRD eGFR. ACEI indicates angiotensin-converting enzyme inhibitor; and LVSF, left ventricular systolic function.

Graphic Jump Location

Tables

Table Graphic Jump LocationTable 2. Data for SUN, Creatinine, and eGFR: Associated Mortality Risks in Patients With MI*
Table Graphic Jump LocationTable 3. Data for SUN, Creatinine, and eGFR: Associated Mortality Risks in Patients With HF*
Table Graphic Jump LocationTable 4. Incremental Adjusted Mortality Risks Associated With Elevated SUN and Decreased Mayo eGFR

References

Smith  GLShlipak  MGHavranek  EP  et al.  Race and renal impairment in heart failure: mortality in blacks versus whites. Circulation 2005;1111270- 1277
Link to Article
Anavekar  NSMcMurray  JJVelazquez  EJ  et al.  Relation between renal dysfunction and cardiovascular outcomes after myocardial infarction. N Engl J Med 2004;3511285- 1295
PubMed Link to Article
Shlipak  MGSmith  GLRathore  SSMassie  BMKrumholz  HM Renal function, digoxin therapy, and heart failure outcomes: evidence from the Digoxin Intervention Group trial. J Am Soc Nephrol 2004;152195- 2203
PubMed Link to Article
Smith  GLVaccarino  VKosiborod  M  et al.  Worsening renal function: what is a clinically meaningful change in creatinine during hospitalization with heart failure? J Card Fail 2003;913- 25
PubMed Link to Article
Dries  DLExner  DVDomanski  MJGreenberg  BStevenson  LW The prognostic implications of renal insufficiency in asymptomatic and symptomatic patients with left ventricular systolic dysfunction. J Am Coll Cardiol 2000;35681- 689
PubMed Link to Article
Hillege  HLGirbes  ARde Kam  PJ  et al.  Renal function, neurohormonal activation, and survival in patients with chronic heart failure. Circulation 2000;102203- 210
PubMed Link to Article
McAlister  FAEzekowitz  JTonelli  MArmstrong  PW Renal insufficiency and heart failure: prognostic and therapeutic implications from a prospective cohort study. Circulation 2004;1091004- 1009
PubMed Link to Article
Al-Ahmad  ARand  WMManjunath  G  et al.  Reduced kidney function and anemia as risk factors for mortality in patients with left ventricular dysfunction. J Am Coll Cardiol 2001;38955- 962
PubMed Link to Article
Shlipak  MGHeidenreich  PANoguchi  HChertow  GMBrowner  WSMcClellan  MB Association of renal insufficiency with treatment and outcomes after myocardial infarction in elderly patients. Ann Intern Med 2002;137555- 562
PubMed Link to Article
Wright  RSReeder  GSHerzog  CA  et al.  Acute myocardial infarction and renal dysfunction: a high-risk combination. Ann Intern Med 2002;137563- 570
PubMed Link to Article
Levey  ASBosch  JPLewis  JBGreene  TRogers  NRoth  DModification of Diet in Renal Disease Study Group, A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Ann Intern Med 1999;130461- 470
PubMed Link to Article
Levey  ASGreene  TKusek  JWBeck  GJ A simplified equation to predict glomerular filtration rate from serum creatinine [abstract]. J Am Soc Nephrol 2000;11155A
 National Kidney Disease Education Program (NKDEP). http://www.nkdep.nih.gov. Accessed March 3, 2005
Sarnak  MJLevey  ASSchoolwerth  AC  et al.  Kidney disease as a risk factor for development of cardiovascular disease: a statement from the American Heart Association Councils on Kidney in Cardiovascular Disease, High Blood Pressure Research, Clinical Cardiology, and Epidemiology and Prevention. Circulation 2003;1082154- 2169
PubMed Link to Article
Hostetter  TH Chronic kidney disease predicts cardiovascular disease. N Engl J Med 2004;3511344- 1346
PubMed Link to Article
Levey  ASCoresh  JBalk  E  et al.  National Kidney Foundation practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Ann Intern Med 2003;139137- 147
PubMed Link to Article
Rule  ADLarson  TSBergstralh  EJSlezak  JMJacobsen  SJCosio  FG Using serum creatinine to estimate glomerular filtration rate: accuracy in good health and in chronic kidney disease. Ann Intern Med 2004;141929- 937
PubMed Link to Article
Foody  JMFerdinand  FDGalusha  D  et al.  Patterns of secondary prevention in older patients undergoing coronary artery bypass grafting during hospitalization for acute myocardial infarction. Circulation 2003;108 ((suppl 1)) II24- II28
PubMed Link to Article
Schemper  M The relative importance of prognostic factors in studies of survival. Stat Med 1993;122377- 2382
PubMed Link to Article
Chobanian  AVBakris  GLBlack  HR  et al.  The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 report. JAMA 2003;2892560- 2572
PubMed Link to Article
Go  ASChertow  GMFan  DMcCulloch  CEHsu  CY Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. N Engl J Med 2004;3511296- 1305
PubMed Link to Article
Shemesh  OGolbetz  HKriss  JPMyers  BD Limitations of creatinine as a filtration marker in glomerulopathic patients. Kidney Int 1985;28830- 838
PubMed Link to Article
Cockcroft  DWGault  MH Prediction of creatinine clearance from serum creatinine. Nephron 1976;1631- 41
PubMed Link to Article
Stevens  LALevey  AS Clinical implications of estimating equations for glomerular filtration rate. Ann Intern Med 2004;141959- 961
PubMed Link to Article
Froissart  MRossert  JJacquot  CPaillard  MHouillier  P Predictive performance of the Modification of Diet in Renal Disease and Cockcroft-Gault equations for estimating renal function. J Am Soc Nephrol 2005;16763- 773
PubMed Link to Article
Smilde  TDHillege  HLNavis  GVoors  AAvan Veldhuisen  DJ Formulas estimating renal function give imprecise estimates in heart failure [abstract]. J Am Coll Cardiol 2005;45172A
Link to Article
Risch  LSagmeister  MHuber  A Chronic renal disease and cardiovascular risk [letter]. N Engl J Med 2005;352199- 200
PubMed Link to Article
Cheng  H Chronic renal disease and cardiovascular risk [letter]. N Engl J Med 2005;352199- 200
PubMed Link to Article
Lamb  EJWebb  MCSimpson  DECoakley  AJNewman  DJO’Riordan  SE Estimation of glomerular filtration rate in older patients with chronic renal insufficiency: is the Modification of Diet in Renal Disease formula an improvement? J Am Geriatr Soc 2003;511012- 1017
PubMed Link to Article
Bostom  AGKronenberg  FRitz  E Predictive performance of renal function equations for patients with chronic kidney disease and normal serum creatinine levels. J Am Soc Nephrol 2002;132140- 2144
PubMed Link to Article
Corsonello  APedone  CIncalzi  RAGemelli  PA Methods for estimating glomerular filtration rate. JAMA 2004;2912819- 2820
PubMed
Maaravi  YBursztyn  MStessman  J The new Mayo Clinic equation for estimating glomerular filtration rate [letter]. Ann Intern Med 2005;142680- 681
PubMed Link to Article
Delanaye  PKrzesinski  JM The new Mayo Clinic equation for estimating glomerular filtration rate [letter]. Ann Intern Med 2005;142679- 680
PubMed Link to Article
Froissart  MCRossert  JHouillier  P The new Mayo Clinic equation for estimating glomerular filtration rate [letter]. Ann Intern Med 2005;142679
PubMed Link to Article
Felker  GMLeimberger  JDCaliff  RM  et al.  Risk stratification after hospitalization for decompensated heart failure. J Card Fail 2004;10460- 466
PubMed Link to Article
Lee  DSAustin  PCRouleau  JLLiu  PPNaimark  DTu  JV Predicting mortality among patients hospitalized for heart failure: derivation and validation of a clinical model. JAMA 2003;2902581- 2587
PubMed Link to Article
Fonarow  GCAdams  KF  JrAbraham  WTYancy  CWBoscardin  WJ Risk stratification for in-hospital mortality in acutely decompensated heart failure: classification and regression tree analysis. JAMA 2005;293572- 580
PubMed Link to Article
Califf  RMPieper  KSLee  KL  et al.  Prediction of 1-year survival after thrombolysis for acute myocardial infarction in the Global Utilization of Streptokinase and TPA for Occluded Coronary Arteries trial. Circulation 2000;1012231- 2238
PubMed Link to Article
Boersma  EPieper  KSSteyerberg  EW  et al. PURSUIT Investigators, Predictors of outcome in patients with acute coronary syndromes without persistent ST-segment elevation: results from an international trial of 9461 patients. Circulation 2000;1012557- 2567
PubMed Link to Article
Antman  EMCohen  MBernink  PJ  et al.  The TIMI risk score for unstable angina/non-ST elevation MI: a method for prognostication and therapeutic decision making. JAMA 2000;284835- 842
PubMed Link to Article
Morrow  DAAntman  EMCharlesworth  A  et al.  TIMI risk score for ST-elevation myocardial infarction: a convenient, bedside, clinical score for risk assessment at presentation: an intravenous nPA for treatment of infarcting myocardium early II trial substudy. Circulation 2000;1022031- 2037
PubMed Link to Article
Jacobs  DR  JrKroenke  CCrow  R  et al.  PREDICT: a simple risk score for clinical severity and long-term prognosis after hospitalization for acute myocardial infarction or unstable angina: the Minnesota heart survey. Circulation 1999;100599- 607
PubMed Link to Article
Krumholz  HMChen  JChen  YTWang  YRadford  MJ Predicting one-year mortality among elderly survivors of hospitalization for an acute myocardial infarction: results from the Cooperative Cardiovascular Project. J Am Coll Cardiol 2001;38453- 459
PubMed Link to Article
Kirtane  AJLeder  DMWaikar  SS  et al.  Serum blood urea nitrogen as an independent marker of subsequent mortality among patients with acute coronary syndromes and normal to mildly reduced glomerular filtration rates. J Am Coll Cardiol 2005;451781- 1786
PubMed Link to Article
Shlipak  MGSarnak  MJKatz  R  et al.  Cystatin-C and risk for mortality and cardiovascular disease in elderly adults. N Engl J Med 2005;3522049- 2060
PubMed Link to Article

Correspondence

CME
Also Meets CME requirements for:
Browse CME for all U.S. States
Accreditation Information
The American Medical Association is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The AMA designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 CreditTM per course. Physicians should claim only the credit commensurate with the extent of their participation in the activity. Physicians who complete the CME course and score at least 80% correct on the quiz are eligible for AMA PRA Category 1 CreditTM.
Note: You must get at least of the answers correct to pass this quiz.
Please click the checkbox indicating that you have read the full article in order to submit your answers.
Your answers have been saved for later.
You have not filled in all the answers to complete this quiz
The following questions were not answered:
Sorry, you have unsuccessfully completed this CME quiz with a score of
The following questions were not answered correctly:
Commitment to Change (optional):
Indicate what change(s) you will implement in your practice, if any, based on this CME course.
Your quiz results:
The filled radio buttons indicate your responses. The preferred responses are highlighted
For CME Course: A Proposed Model for Initial Assessment and Management of Acute Heart Failure Syndromes
Indicate what changes(s) you will implement in your practice, if any, based on this CME course.

Multimedia

Some tools below are only available to our subscribers or users with an online account.

2,147 Views
57 Citations
×

Related Content

Customize your page view by dragging & repositioning the boxes below.

Articles Related By Topic
Related Collections
PubMed Articles
Jobs
JAMAevidence.com

Users' Guides to the Medical Literature: A Manual for Evidence-Based Clinical Practice, 3rd ed
Is There Potentially Compelling Evidence for a Class Effect?

Users' Guides to the Medical Literature: A Manual for Evidence-Based Clinical Practice, 3rd ed
Are the End Points in Randomized Clinical Trials Important to Patients?