4Medical Review

The Association of Blood Urea Nitrogen Levels and

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

Coronary Artery Disease

Robert Ostfeld,* Michael Spinelli,* Disha Mookherjee,* Dvorah Holtzman,* Abu Shoyeb,† Michael Schaefer,*
Thomas Kawano,* Sanjay Doddamani,* Daniel Spevack,* Yunling Duc‡

*Division of Cardiovascular Medicine, Department of Medicine

Albert Einstein College of Medicine, Montefiore Medical Center
Bronx, New York 10461

†Memorial Sloan-Kettering Cancer Center

New York, New York 10065

‡Divisionof Biostatistics, Department of Epidemiology & Population Health

Albert Einstein College of Medicine
Bronx, New York 10461

ABSTRACT associated with the burden of CAD on cardiac catheter-

ization and with ruling in for myocardial infarction (MI).
Renal dysfunction has been associated with adverse
cardiovascular outcomes. Estimates of renal function
routinely utilize creatinine-based measures. Serum METHODS
blood urea nitrogen (BUN) levels, however, may provide
supplemental information in regard to renal function We performed a retrospective chart review on 156 con-
as renal proximal tubule cells may increase BUN re- secutive adult patients presenting to the Montefiore
absorption in the setting of increased neurohormonal Medical Center Emergency Department beginning on
activation. We performed a retrospective chart review January 1, 2002, with symptoms of unstable angina and
on 156 consecutive adult patients presenting to the no known prior history of CAD who underwent cardiac
Montefiore Medical Center Emergency Department catheterization as part of their index hospitalization. 
with symptoms of unstable angina and no known prior No known CAD was defined as: no known history of MI,
history of coronary artery disease (CAD) who underwent percutaneous coronary intervention, or coronary artery
cardiac catheterization as part of their index hospital- bypass surgery. Additional exclusion criteria were: (1)
ization. On multivariate analysis, admission serum BUN cardiac catheterization within the previous six months,
was associated with an increased burden of CAD on (2) serum creatinine > 3.0 mg/dl or hemodialysis, (3)
cardiac catheterization and was not associated with a history of heart failure, or (4) other severe acute ill-
ruling in for myocardial infarction (MI). ness or organ failure. Admission history and laboratory
data were obtained. Three patients were missing car-
diac enzymes. Creatinine clearance (CrCl) was calculated
INTRODUCTION with the Cockcroft-Gault equation. Hyperlipidemia was
defined as a history of hyperlipidemia or anti-hyperlip-
Renal dysfunction has been associated with adverse car- idemic medication use.  Diabetes was defined as a his-
diovascular outcomes (McCullough, 2003; Dossetor, 1966; tory of diabetes or anti-hyperglycemic medication use.
Conte et al., 1987; Aronson et al., 2004). In these studies Smoking was defined as any prior history of tobacco
and others, renal function has routinely been assessed smoking. Hypertension (HTN) was defined as a history
with an estimated creatinine clearance, serum creati- of HTN or anti-hypertensive medication use. Positive
nine, or an estimated glomerular filtration rate derived cardiac markers defined ruling in for MI. 
from the serum creatinine (Suwaidi et al., 2002). 
A priori, we developed a burden of CAD score, with
Serum blood urea nitrogen (BUN) levels, however, may higher values representing a greater burden of CAD, as
provide supplemental information in regard to renal described previously (Usberti et al., 1985). We scored
function as renal proximal tubule cells may increase BUN the burden of CAD as follows: (1) left main coronary
reabsorption in the setting of increased neurohormonal artery; each individual stenosis of 20-49% was given a
activation (Shlipak et al., 2002). Accordingly, higher score of 0.5, 50-69%, 1.0, and >=70%, 1.5, respectively,
serum BUN has been associated with adverse outcomes (2) left anterior descending (LAD) artery system (which
in subjects with acute coronary syndromes (Levey et al., included the LAD and diagonals); each individual steno-
1999). We examined whether admission serum BUN in sis of 20-49% was given a score of 0.5, 50-69%, 1.0, and
subjects presenting with symptoms of unstable angina >=70%, 1.5, respectively, (3) left circumflex (LCx) system
and without known coronary artery disease (CAD) was (which included the LCx and obtuse marginals); each

The Einstein Journal of Biology and Medicine 3

4Medical Review
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
The Association of Blood Urea Nitrogen Levels and Coronary Artery Disease

individual stenosis of 20-49% was given Table 1: Baseline Patient Characteristics

a score of 0.5, 50-69%, 1.0, and >=70%,
1.5, respectively, and (4) right coronary Overall* (n=156)
artery (RCA) system (which included the Age 63.2 (+/-) 13.3
RCA, posterior descending artery and
posterior left ventricular branches); each Male 52.6%
individual stenosis of 20-49% was given a
score of 0.5, 50-69%, 1.0, and >=70%, 1.5, Creatinine Clearance (mL/min) 94.0 (+/- 59.7)
respectively.  If a ramus branch was pres- Diabetes 28.2%
ent, each individual stenosis of 20-49%
was given a score of 0.25, 50-69%, 0.5, BMI (kg/m2) 28.3% (+/- 5.9%)
and >=70%, 0.75. The score for each indi-
HTN 73.1%
vidual lesion was totaled into an overall
burden of CAD score for each patient. Dyslipidemia 41.6% (2 missing)
The percent stenoses were obtained from
official catheterization reports from one Smoking 33.8% (11 missing)
of two experienced interventional cardi-
BUN (mg/dl) 17.4 (+/- 6.0)
ologists. The scoring system was consid-
ered “appropriate” by three independent CAD Burden Score 3.9% (+/- 4.1)
cardiologists. The CAD burden score was
dichotomized at the 75th percentile. *Three patients are missing cardiac enzymes
BMI = body mass index; BUN = blood urea nitrogen; CAD = coronary artery disease;
HTN = hypertension
T-tests and chi square analyses were per-
5 
formed when comparing means and
proportions of baseline measurements 1.09 (1.01, 1.12); p=0.02). When MI status during index
between subjects with MI and without MI, and between hospitalization was added to the multivariate model,
subjects with CAD burdens above and below the 75th the association of BUN and burden of CAD remained
percentile, respectively. To evaluate the relationships significant (p=0.05). 
between serum BUN and CAD burden, and between
serum BUN and MI, univariate analyses and multivari- Baseline characteristics for those who did and did not
ate analyses adjusting for baseline characteristics were rule in for MI are presented in Table 3. Those subjects
conducted using logistic regression models.  All statisti- who ruled in for myocardial infarction had a lower body
cal analysis was performed with the MDAS Version 2.0 mass index (BMI) (27.0 mg/dl vs. 29.3 mg/dl, p=0.02), and
(eSKay Software TM 2004) statistical software package. a higher CAD burden score, (5.2 vs. 2.9, p<0.01), com-
The Montefiore Medical Center Institutional Review pared with those who did not rule in for myocardial
Board approved this study. infarction. 

On univariate and multivariate analysis, an increased

RESULTS BUN was not associated with an increased odds of rul-
ing in for MI (OR 0.99 (0.07, 1.02); p=0.92) and (OR 0.99
Baseline characteristics for the entire cohort are pre- (0.96, 1.05); p=0.82), respectively. 
sented in Table 1. Baseline characteristics for the dichot-
omized CAD burden variable are presented in Table
2.  The CAD burden score range was 0-19 with a 75th DISCUSSION
percentile of seven.  Subjects with a CAD burden score
greater than or equal to the 75th percentile were older We found that elevated serum BUN on admission was
(69.2 years vs. 61.0, p=0.04), had lower CrCl (75.7 vs. associated with an increased burden of CAD on cardiac
100.3, p<0.01), lower BMI (26.5 vs. 28.9, p<0.01), and catheterization during index hospitalization in patients
higher BUN (20.5 vs. 16.5, p<0.01).  who presented with symptoms of unstable angina and
without known cardiovascular disease. To the best of
On univariate analysis, each 1 mg/dl increase in BUN was our knowledge, this study is the first to demonstrate
associated with an average increased odds of having a this association. Furthermore, each 1mg/dl increase in
CAD burden score greater or equal to the 75th percen- BUN was associated with an increased burden of CAD. 
tile of 12% (OR 1.12 (1.05, 1.19); p<0.01).  On multivari- Admission BUN, however, was not associated with ruling
ate analysis, which included age, sex, CrCl, body mass in for myocardial infarction. 
index, history of smoking, hyperlipidemia, HTN, and
diabetes, each 1 mg/dl increase in BUN was associated Blood urea nitrogen may have pro-atherosclerotic
with an average increased odds of having a CAD burden effects, as uremia has been associated with an increased
score greater or equal to the 75th percentile of 9% (OR burden of oxidative stress (Himmelfarb et al., 2002).

4 EJBM, Copyright © 2010

 4Medical Review
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
The Association of Blood Urea Nitrogen Levels and Coronary Artery Disease
4; 351:128512
Table 2: Baseline Patient Characteristics by Dichotomized CAD Burden Score
CAD Score < 75% (n=116) ≥75% (n=40) p value*

Age 61.0 (+/- 13.5) 69.2 (+/- 10.9) 0.04

Male 50.9% 50.0% NS

Creatinine clearance (mL/min) 100.3 (+/- 65.4) 75.7 (+/- 33.1) <0.01

Diabetes 25.7% 35.0% NS

BMI (kg/m2) 28.9 (+/- 6.4) 26.5 (+/- 4.1) <0.01

HTN 71.0% 80.0% NS

Dyslipidemia 39.4% (2 missing) 47.5% NS

Smoking 33.0% (7 missing) 36.1% (4 missing) NS

BUN (mg/dl) 16.5 (+/- 5.2) 20.5 (+/- 7.2) <0.01

*A p value of <0.05 was considered significant

BMI = body mass index; BUN = blood urea nitrogen; CAD = coronary artery disease;
HTN = hypertension; NS = not significant

BUN may also both inhibit nitric oxide synthesis and pro- associated with both an increased burden of CAD and
mote macrophage proliferation (Moeslinger et al., 1999). with ruling in for myocardial infarction. This finding may
Specifically, in vivo studies demonstrate that increas- be secondary to chance. Alternatively, it may lend support
ing levels of urea inhibit nitric oxide synthesis in mouse to the “obesity paradox,” where overweight or obese
macrophages with concurrent macrophage proliferation patients have greater cardiovascular risk yet appear to
(Conte et al., 1987). Furthermore, uremia accelerates ath- have improved outcomes compared to patients who are
erosclerosis in Apolipoprotein E-deficient mice.   not overweight or obese (Pingitore et al., 2007).

Other studies indicate that uremia induces the expression Our study has several limitations. It is a retrospective
of osteoblast differentiation factor Cbfa1 in the intima analysis.  The BUN levels were a single measurement
and media of arteries, which may lead to vascular calcifi- and are not necessarily a reflection of one’s chronic
cation (Moe et al., 2003). Elevated BUN may also serve as level. BUN elevation may occur for reasons other than
a marker of an activated sympathetic nervous system and/ neurohormonal up-regulation and/or the presence of
or an upregulated renin-angiotensin system, reported renal dysfunction, such as bleeding, diuretic or steroid
promoters of atherosclerosis (Kirtane et al., 2005; Ostfeld use and/or dietary sources (Mark et al., 1994). No patient
et al., 2006; Manuck et al., 1988; Rozanski et al., 1990; in our study, however, had known active bleeding, or
Tummala et al., 1999). Activation of these neurohor- was known to be taking loop diuretics or steroids. We
monal systems has been associated with increased BUN were unable to account for potential dietary influences
reabsorption in the renal tubules. on serum BUN. Furthermore, our measure of the bur-
den of CAD, as previously reported (Usberti et al., 1985),
We observed this association between increased BUN although deemed “appropriate” by three independent
and CAD while correcting for creatinine clearance.  cardiologists, has not been validated.  However, previous
Concurrently, Kirtane et al. (2005) reported that increas- studies have utilized a similar scale in grading coronary
ing BUN predicts poor outcome in subjects with acute artery lesions based on angiographic findings (Warnholtz
coronary syndromes despite normal or mildly reduced et al., 1999; Goetz, 1997).
glomerular filtration rates. Consequently, an elevated
serum BUN may represent an independent marker of
renal dysfunction, which would further support the well- CONCLUSIONS
established association between renal disease and CAD
(Anvekar et al., 2004; Foley et al., 2005). However, addi- In conclusion, our study suggests that an elevated serum
tional study is required to evaluate this point.  BUN in subjects presenting with signs and symptoms of
unstable angina and without known CAD may predict a
Counterintuitively, we found that a reduced BMI was larger burden of CAD on cardiac catheterization inde-

The Einstein Journal of Biology and Medicine 5

4Medical Review
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
The Association of Blood Urea Nitrogen Levels and Coronary Artery Disease

Table 3: Baseline Patient Characteristics by Presence or Absence of MI*

†
MI (+) (n=68) MI (-) (n=85) p value

Age 63.4 (+/- 12.9) 63.0 (+/- 13.9) NS

Male 50.0% 55.3% NS

Creatinine clearance (mL/min) 95.5 (+/- 75.4) 93.9 (+/- 44.7) NS

Diabetes 20.6% 32.9% NS

2
BMI (kg/m ) 27.0 (+/- 5.1) 29.3 (+/- 6.4) 0.02

HTN 67.6% 76.5% NS

Dyslipidemia 37.3% (1 missing) 42.9% (1 missing) NS

Smoking 36.7% (8 missing) 30.5% (3 missing) NS

BUN (mg/dl) 17.4 (+/- 6.1) 17.5 (+/- 6.0) NS

CAD burden score 5.2 (+/- 3.9) 2.9 (+/- 4.0) <0.01

*Three patients are missing cardiac enzymes

†A p value of <0.05 was considered significant
BMI = body mass index; BUN = blood urea nitrogen; CAD = coronary artery disease
HTN = hypertension; MI = myocardial infarction; NS = not significant

pendent of creatinine clearance. Further study is war-

Levey, A.S., Bosch, J.P., Lewis, J.B., Greene, T., Rogers, N., Roth, D. (1999). A
ranted to explore this association.  more accurate method to estimate glomerular filtration rate from serum crea-
tinine: A new prediction equation. Ann Intern Med 130:461-70.

Manuck, S.B., Kaplan, J.R., Adams, M.R., Clarkson, T.B. (1988). Effects of stress
REFERENCES and the sympathetic nervous system on coronary atherosclerosis in the cyno-
molgus macaque. Am Heart J 116:328-33.
Anvekar, N.S., McMurray, J.J., Velazquez, E.J., Solomon, S.D., Kober, L., Roulau,
J.L., White, H.D., Nrdlander, R., Maggioni, A., Dickstein, K., Zelenkfske, S., Mark, D.B., Nelson, C.L., Califf, R.M., Harrell Jr., F.E., Lee, K.L., Jones, R.H.,
Leimberger, J.D., Califf, R.M., Pfeffer, M.A. (2004). Relation between renal Fortin, D.F., Stack, R.S., Glower, D.D., Smith, L.R., DeLong, E.R., Smith, P.K.,
dysfunction and cardiovascular outcomes after myocardial infarction. N Engl Reves, J.G., Jollis, J.G., Tcheng, J.E., Muhlbaier, L.H., Lowe, J.E., Phillips, H.R.,
J Med 351(13):1285-95. Pryor, D.B. (1994). Continuing evolution of therapy for coronary artery dis-
ease. Circulation 89:2015-25.
Aronson, D., Mittleman, M.A., Burger, A.J. (2004). Elevated blood urea nitro-
gen level as a predictor of mortality in patients admitted for decompensated McCullough, P.A. (2003). Interface between heart disease and renal dysfunc-
heart failure. Am J Med 116:466-73. tion: From association to action. ACC Curr J Rev 12:20-4.

Conte, G., Dal Canton, A., Terribile, M., Cianciaruso, B., Di Minno, G., Pannain, Moe, S., Duan, D., Doehle, D.P., O’Neill, K.D., Chen, N.S. (2003). Uremia induces
M., Russo, D., Andreucci, V.E. (1987). Renal handling of urea in subjects with the osteoblast differentiation factor Cbfa1 in human blood vessels. Kidney
persistent azotemia and normal renal function. Kidney Int 32:721-7. Int 63:1003-11.

Dossetor, J.B. (1966). Creatinemia versus uremia: The relative significance of Moeslinger, T., Roswitha, F., Volf, I., Brunner, M., Baran, H., Koller, E.,
blood urea nitrogen and serum creatinine concentration in azotemia. Ann Spieckermann, P. (1999). Urea induce macrophage proliferation by inhibition
Intern Med 65:1287-99. of inducible nitric oxide synthesis. Kidney Int 56:581-8.

Foley, R.N., Wang, C., Collins, A.J. (2005). Cardiovascular risk factor profiles Ostfeld, R., Mookherjee, D., Spinelli, M., Holtzman, D., Shoyeb, A., Schaefer,
and kidney function stage in the US general population: The NHANES III study. M., Doddamani, S., Spevack, D., Du, Y. (2006). A triglyceride/high-density lipo-
Mayo Clin Proc 1980(10):1270-7.  protein ratio > or = 3.5 is associated with an increased burden of coronary
artery disease on cardiac catheterization. JCMS 1:13-5.
Goetz, R.M. (1997). The antiatherogenic potential of blocking the renin-an-
giotensin system. Expert Opin Investig Drugs 6:1651-8. Pingitore, A., Di Bella, G., Lombardi, N., Iervasi, G., Strata, E., Aquaro, G.D.,
Positano, V., De Marchi, D., Rossi, G., L’Abbate, A., Rovai, D. (2007). The obesity
Himmelfarb, J., Stenvinkel, J., Ikizler, T.A., Hakim, R.M. (2002). The elephant paradox and myocardial infarct size. J Cardiovasc Med 8(9):714-7.
in uremia: Oxidant stress as a unifying concept of cardiovascular disease in
uremia. Kidney Int 62:1524-38. Rozanski, A., Blumenthal, J.A., Kaplan, J. (1990). Impact of psychological fac-
tors on the pathogenesis of cardiovascular disease and implications for ther-
Kirtane, A.J., Leder, D.M., Waikar, S.S., Chertow, G.M., Ray, K.K., Pinto, D.S., apy. Circulation 99:2192-2217.
Karmpaliotis, D., Burger, A.J., Murphy, S.A., Cannon, C.P., Braunwald, E.,
Gibson, C.M. (2005). Serum blood urea nitrogen as an independent marker Shlipak, M.G., Heidenreich, P.A., Noguchi, H., Chertow, G.M., Browner, W.S.,
of subsequent mortality among patients with acute coronary syndromes McClellan, M.B. (2002). Association in renal insufficiency with treatment and
and normal to mildly reduced glomerular filtration rates. J Am Coll Cardiol outcomes after myocardial infarction in elderly patients. Ann Intern Med
45:1781-86. 137:555-62.

6 EJBM, Copyright © 2010

 4Medical Review
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
The Association of Blood Urea Nitrogen Levels and Coronary Artery Disease

Suwaidi, J.A., Reddan, D.N., Williams, K., Pieper, K.S., Harrington, R.A., Califf, Usberti, M., Federico, S., Di Minno, G., Ungaro, B., Ardillo, G., Pecoraro, C.,
R.M., Granger, C.B., Ohman, E.M., Holmes, D.R. (2002). Prognostic implications Cianciaruso, B., Cerbone, A.M., Cirillo, F., Pannain, M., Gargiulo, A., Andreucci,
of abnormalities in renal function in patients with acute coronary syndromes. V.E. (1985). Effects of angiotensin II on plasma ADH, prostaglandin synthesis,
Circulation 106:974-80. and water excretion in normal humans. Am J Physiol 248:F254-9.

Tummala, P.E., Chen, X.L., Sundell, C.L., Laursen, J.B., Hammes, P., Alexander, Warnholtz, A., Nickenig, G., Schulz, E., Macharzina, R., Brasen, J.H., Skatchkov,
R.W., Harrison, D.G., Medford, R.M. (1999). Angiotensin II induces vascular cell M., Heitzer, T., Stasch, J.P., Griendling, K.K., Harrison, D.G., Bohm, M., Meinertz,
adhesion molecule-1 expression in rat vasculature. Circulation 100:1223-9. T., Munzel, T. (1999). Increased NADH-oxidase-mediated superoxide produc-
tion in the early stages of atherosclerosis. Evidence for involvement of the
renin-angiotensin system. Circulation 99:2027-33.
5

The Einstein Journal of Biology and Medicine 7