ORIGINAL ARTICLE

Outcomes of Percutaneous Coronary

Interventions in Patients With Anemia
Presenting With Acute Coronary Syndrome
Mohammed A. Al-Hijji, MD; Rajiv Gulati, MD, PhD; Ryan J. Lennon, MS;
Malcolm Bell, MD; Abdallah El Sabbagh, MD; Jae Yoon Park, MD; Joshua Slusser, BS;
Gurpreet S. Sandhu, MD, PhD; Guy S. Reeder, MD; Charanjit S. Rihal, MD, MBA;
and Mandeep Singh, MD, MPH

Abstract

Objective: To study the influence of anemia on long-term outcomes of patients with acute coronary
syndrome undergoing percutaneous coronary intervention (PCI).
Patients and Methods: The study included 5668 consecutive unique patients with acute coronary
syndrome who underwent PCI at Mayo Clinic from January 1, 2004, through December 31, 2014. The
patients were stratified on the basis of the presence (hemoglobin [Hgb] level, <13 g/dL in men and <12
g/dL in women) and severity (moderate to severe Hgb level, <11 g/dL in men and women) of pre-PCI
anemia and compared with patients without anemia. The primary outcomes were in-hospital and long-
term all-cause mortality after balancing baseline comorbidities using the inverse propensity weighting
method.
Results: Unadjusted all-cause in-hospital mortality (4.6% [84 of 1831] vs 2.0% [75 of 3837]) and 5-year
follow-up mortality (44.4% [509] vs 15.4% [323]) were higher in patients with anemia than in those
without anemia (P<.001 for both). After applying inverse propensity weighting analysis, the all-cause in-
hospital mortality (2.0% [37] vs 2.0% [75]; P¼.85) and 5-year mortality (17.8% [203] vs 15.4% [323];
P¼.05) were not significantly different between patients with and without anemia; however, there were
higher rates of all-cause 5-year mortality in patients with moderate to severe anemia (22.3% [113] vs
15.4% [323]; P<.001) compared with patients without anemia. The trend in 5-year mortality was driven
by increased noncardiac mortality in patients with anemia (10.2% [91] vs 7.1% [148]; P¼.04) and
moderate to severe anemia (10.4% [52] vs 7.1% [148]; P¼.006) when compared with nonanemic patients.
Conclusion: After accounting for differences in risk profiles of anemic and nonanemic patients, anemia
appeared to be an independent risk factor for increased long-term all-cause and noncardiac mortality.
ª 2018 Mayo Foundation for Medical Education and Research n Mayo Clin Proc. 2018;93(10):1448-1461

From the Division of Cardio-
vascular Diseases (M.A.A.-H.,
R.G., M.B., A.E.-S., J.Y.P., G.S.S.,
G.S.R., C.S.R., M.S.) and Divi-
sion of Biomedical Statistics
and Informatics (R.J.L., J.S.),
Mayo Clinic, Rochester, MN.
T
he prevalence of anemia in patients
with acute coronary syndrome (ACS)
is high, ranging between 10% and
32%.1-4 More so, multiple registries5-11 and
post hoc analyses of randomized controlled
trials (RCTs)3,4,12-14 have reported that the
presence of anemia in ACS is associated with
worse clinical outcomes including death,
ischemic events, and heart failure. In their pi-
lot randomized controlled trial, Cooper et al15
found that treatment of anemia with liberal
blood transfusion in ACS was associated with
a 75% relative risk increase in cumulative in-
hospital death, acute heart failure, and
recurrent infarction. Purported mechanisms
for poor prognosis include reduced myocar-
dial oxygen supply, increased myocardial oxy-
gen demand and ischemia, volume expansion,
and increased bleeding risk.4,16-18 Overt
bleeding and need for blood transfusion in
the setting of ACS are also more prevalent in
anemic patients and associated with higher
mortality and recurrent infarction.1,18-23 The
causal role of anemia, however, is confounded
by coexistent medical comorbidities.3,4,10,24-26
leading to the lack of firm recommendations
on its management in the most recent ACS
guidelines.1,2

1448 Mayo Clin Proc. n October 2018;93(10):1448-1461 n https://doi.org/10.1016/j.mayocp.2018.03.030

www.mayoclinicproceedings.org n ª 2018 Mayo Foundation for Medical Education and Research
BASELINE ANEMIA AND PCI OUTCOMES
The need to reexamine anemia as a risk
marker or independent risk factor for patients
with coronary artery disease is underscored by
changing demographic characteristics toward
the elderly with higher prevalence of comorbid-
ities (eg, malignancy, diabetes, gastrointestinal
tract disorders, and chronic renal disease), a
population seldom studied in previous
randomized controlled trials.3,4,12-14 At Mayo
Clinic, detailed records of the presence and
severity of anemia and long-term data,
including mortality, cause of death, and cardio-
vascular ischemic events, on all patients who
undergo percutaneous coronary intervention
(PCI) are available. An inverse propensity
weighting (IPW) method was utilized for this
study to account for extensive differences in
baseline risk profiles of patients with and
without anemia. With this background, we
hypothesized that in patients presenting with
ACS and undergoing PCI, anemia (and its
severity) is associated with poor in-hospital
and long-term outcomes, following adjustment
by propensity scores.
PATIENTS AND METHODS
Data Collection
At the time of PCI, patient-specific data were
entered into a prospective clinical registry.27,28
The Mayo Clinic PCI registry includes clinical,
procedural, and angiographic data on all
patients undergoing PCI. Patients were con-
tacted at 6 months, 12 months, and yearly
thereafter by a clinical research nurse. Medical
records of all patients requiring hospitalization
at Mayo Clinic, or elsewhere, were reviewed to
further characterize any clinical events during
follow-up. Separate approval for the current
study was obtained from the Mayo Clinic Insti-
tutional Review Board. Patients who denied
research authorization were excluded from the
study in accordance with Minnesota law.
Study Population
Consecutive patients with ACS who underwent
PCI from January 1, 2004, through December,
31, 2014, were stratified on the basis of the
presence (hemoglobin [Hgb] level, <13 g/dL in
men and <12 g/dL in women; to convert
values to g/L, multiply by 10.0) and severity
(moderate to severe anemia Hgb level, <11 g/
dL in men and women) of anemia detected on
Mayo Clin Proc. n October 2018;93(10):1448-1461
www.mayoclinicproceedings.org
hospital admission, before PCI, using the World
Health Organization definition.29 Baseline
demographic characteristics; comorbid condi-
tions defined by the Sachdev index (includes
cigarette smoking, hypertension, diabetes melli-
tus, cerebrovascular accidents, chronic kidney
disease, and malignancy and previously found
to be associated with poor survival in patients
with coronary artery disease)30; Mayo Clinic
risk score (MCRS) for adverse cardiovascular
events27,28; and type of ACS (unstable angina,
noneST-elevation myocardial infarction
[NSTEMI] or ST-elevation myocardial infarction
[STEMI]) presentations were extracted from the
Mayo Clinic PCI registry. Angiographic data
including the type, number, and locations of
lesions as well as the type of stents deployed
were compared between the 2 groups. The use
of ACS guideline-directed medical therapy,
including the use of dual antiplatelet therapy
and glycoprotein IIb/IIIa inhibitor, was noted.1,2
Clinical Outcomes
The primary outcomes were in-hospital and
5-year cause-specific (cardiac and noncardiac)
mortality. Secondary outcomes were
in-hospital and 5-year major adverse cardiac
events (MACE) (composite of all-cause death,
myocardial infarction [MI], stroke, or repeated
revascularization), in-hospital bleeding, need
for blood transfusion (1 U), and number of
units transfused. Bleeding was defined as
reported site hematoma (documented in med-
ical record and required transfusion, surgery,
or prolonged hospital stay), femoral bleeding
(excessive bleeding from the femoral artery
requiring treatment), gastrointestinal tract
bleeding (hematemesis, melena, or hemato-
chezia), cerebrovascular bleeding (docu-
mented radiographically), and retroperitoneal
bleeding (bleeding postcatheterization into
retroperitoneal space on the side of the punc-
ture site, evidenced by flank pain and tender-
ness to palpation and ecchymosis on abdomen
and flank, and documented by computed
tomography or ultrasonography). In-hospital
acute kidney injury was defined as documen-
tation of “acute renal failure or acute kidney
injury” of any severity as a complication in
the medical record. Cause of death was further
divided into cardiac and noncardiac using
medical record review and death certificates
of all patients and information from the
n https://doi.org/10.1016/j.mayocp.2018.03.030 1449
 MAYO CLINIC PROCEEDINGS

TABLE 1. Baseline Characteristics and Clinical Presentations of the 5668 Study Patientsa,b
Anemiac No anemiac
Variable (n¼1831) (n¼3837) P value
Demographic characteristics
Age (y) 72.511.5 64.812.7 <.001
Male 1175 (64.2) 2795 (72.8) <.001
Body mass index (kg/m2) 29.96.8 30.55.8 .001
Comorbidities
Diabetes mellitus 745/1824 (40.8) 823/3826 (21.5) <.001
Chronic kidney disease 180/1802 (10.0) 43/3819 (1.1) <.001
Hypertension 1600/1801 (88.8) 2760/3719 (74.2) <.001
Chronic heart failure 603/1792 (33.6) 395/3790 (10.4) <.001
Hyperlipidemia 1552/1785 (86.9) 2866/3638 (78.8) <.001
Current smoker 243/1787 (13.6) 998/3767 (26.5) <.001
Previous MI 591/1823 (32.4) 763/3814 (20.0) <.001
Previous CABG 492/1829 (26.9) 549/3834 (14.3) <.001
Previous PCI 662/1831 (36.2) 1029/3837 (26.8) <.001
Anticoagulation therapy 99/1827 (5.4) 84/3830 (2.2) <.001
Malignancy 371/1828 (20.3) 378/3832 (9.9) <.001
History of stroke 306/1816 (16.9) 335/3815 (8.8) <.001
Sachdev index 4.23.8 2.02.1 <.001
Clinical presentation
Acute heart failure 474 (25.9) 291 (7.6) <.001
LVEF (%) <.001
>40 768 (41.9) 1448 (37.7)
40 297 (16.2) 317 (8.3)
Unstable angina 588 (32.1) 1557 (40.6) <.001
NSTEMI 830 (45.3) 1187 (30.9) <.001
STEMI 413 (22.6) 1093 (28.5) <.001
Estimated risk of in-hospital MACE (%)d 3.5 (2.3, 6.7) 2.3 (1.4, 3.5) <.001
Estimated risk of in-hospital, death (%)d 0.8 (0.4, 2.2) 0.4 (0.2, 0.8) <.001
GFR at admission (mL/min/1.73 m2)e <.001
>90 300 (16.8) 1078 (28.9)
9060 663 (37.0) 2005 (53.7)
60-30 643 (35.9) 621 (16.6)
<30-15 109 (6.1) 24 (0.6)
<15 75 (4.2) 7 (0.2)
a
CABG ¼ coronary artery bypass graft; GFR ¼ glomerular filtration rate; LVEF ¼ left ventricular ejection fraction; MACE ¼ major adverse
cardiac events; MI ¼ myocardial infarction; NSTEMI ¼ noneST-elevation MI ; PCI ¼ percutaneous coronary intervention; STEMI ¼ ST-
elevation MI.
b
Data are presented as mean  SD, No. (percentage) of patients, or median (25th, 75th percentiles).
c
Anemia was defined as a hemoglobin level of <13 g/dL in men and <12 g/dL in women (to convert values to g/L, multiply by 10.0). No
anemia was defined as a hemoglobin level of 13 g/dL in men and 12 g/dL in women.
d
Estimated using Mayo Clinic risk score.
e
GFR was not available in 41 patients with and 102 patients without anemia (percentages are based on 1790 and 3735 patients,
respectively).

National Death Index if the cause of death Statistical Analyses

could not initially be determined.31,32 Cardiac
death was defined as mortality due to MI,
cardiac arrest or arrhythmia, congestive heart
failure, structural heart disease, or procedure-
related death. All other causes of death were
labeled as noncardiac death as previously
described.31,32
n n
1450 Mayo Clin Proc.
Continuous variables are presented as mean 
SD or median (25th, 75th percentiles).
Discrete variables are summarized as fre-
quency (percentage). Comparisons of baseline,
procedural, and angiographic characteristics
between groups were made using the Pearson
c2 test for categorical variables and the
October 2018;93(10):1448-1461 https://doi.org/10.1016/j.mayocp.2018.03.030
www.mayoclinicproceedings.org
BASELINE ANEMIA AND PCI OUTCOMES

TABLE 2. Angiographic Results and Managementa,b

Variable Anemiac (n¼1831) No anemiac (n¼3837) P value
Angiographic findings d

No. of diseased vessels .003

1 1479 (82.3) 3224 (85.9)
2 294 (16.4) 496 (13.2)
3 23 (1.3) 34 (0.9)
Left main coronary artery 141 (7.8) 99 (2.6) <.001
Management
Type of stente
BMS 484 (28.8) 636 (17.9) <.001
DES 1195 (71.2) 2915 (82.0) <.001
Femoral accessf 1512 (84.4) 3004 (80.2) <.001
Radial accessf 294 (16.4) 757 (20.2) <.001
Aspirin 1795 (98.0) 3756 (97.9) .72
Plavix 1716 (93.7) 3612 (94.1) .54
Ticagrelor 55 (3.0) 194 (5.1) <.001
Heparin 1803 (98.5) 3795 (98.9) .17
Bivalirudin 22 (1.2) 32 (0.8) .18
Glycoprotein IIb/IIa 940 (51.3) 2481 (64.7) <.001
b-Blocker 1665 (90.9) 3438 (89.6) .12
ACEI/ARB 1077 (58.8) 2229 (58.1) .60
Statin 1613 (88.1) 3398 (88.6) .67
a
ACEI ¼ angiotensin-converting enzyme inhibitor; ARB ¼ angiotensin receptor blocker; BMS ¼ bare metal stent; DES ¼ drug-eluting
stent.
b
Data are presented as No. (percentage) of patients.
c
Anemia was defined as a hemoglobin level of <13 g/dL in men and <12 g/dL in women (to convert values to g/L, multiply by 10.0). No
anemia was defined as a hemoglobin level of 13 g/dL in men and 12 g/dL in women.
d
Angiographic findings were not available in 35 patients with and 83 patients without anemia (percentages are based on 1796 and 3754
patients, respectively).
e
Type of stent was not available in 152 patients with and 286 patients without anemia (percentages are based on 1679 and 3551 patients,
respectively).
f
0.8% of cases (40/5536) were crossovers between radial and femoral access. Data were not available in 40 patients with and 92 patients
without anemia (percentages are based on 1791 and 3745 patients, respectively). Note: the numbers add to 5567, not 5667. There
were also 11 patients who did not have either radial or femoral access (presumably brachial was used).

Student t test or rank-sum test for continuous
variables. The primary outcome (all-cause
long-term mortality) and secondary outcome
(long-term MACE) were computed using the
Kaplan-Meier method, with comparisons
made via the log-rank test. Cause-specific
mortality estimates were calculated using cu-
mulative incidence methods to account for
competing risks, with comparisons made via
the Fine and Grey test. The IPW method
was utilized to preserve the number of patients
in the study and reduce the effect of the exten-
sive systemic differences in comorbidities be-
tween groups.33,34 We included 33 variables
obtained from baseline comorbidities, clinical
presentation, and angiographic and manage-
ment characteristics in the propensity score
model, with continuous variables modeled us-
ing restricted cubic splines and interactions
included as necessary to achieve covariate
balance. Control patients without anemia
were assigned a weight of 1, while patients
with anemia were weighted to achieve a risk
profile similar to the nonanemic group. A
similar method was applied when comparing
patients with moderate to severe anemia and
nonanemic controls. After applying the pro-
pensity method, excellent balance was
achieved between patients with and without
anemia (Supplemental Table 1, available on-
line at http://www.mayoclinicproceedings.
org) and patients with moderate to severe ane-
mia and without anemia (Supplemental
Table 2, available online at http://www.
mayoclinicproceedings.org). After balancing,
categorical variables and continuous variables
were compared using absolute standardized
differences. As a secondary analysis, multivar-
iate modeling was used to estimate the effect
of anemia on outcomes. Logistic regression

Mayo Clin Proc. n October 2018;93(10):1448-1461 n https://doi.org/10.1016/j.mayocp.2018.03.030 1451

www.mayoclinicproceedings.org
 MAYO CLINIC PROCEEDINGS

TABLE 3. Unadjusted In-Hospital Outcomesa

Variable Anemiab (n¼1831) No anemiab (n¼3837) P value
Primary Outcomes
All-cause mortality 84 (4.6) 75 (2.0) <.001
Cause of in-hospital death
Cardiac 70 (3.8) 65 (1.7) <.001
Noncardiac 14 (0.8) 10 (0.3) .006
Major adverse cardiovascular events 104 (5.7) 125 (3.3) <.001
Secondary outcomes
Recurrent revascularization 17 (0.9) 49 (1.3) .25
Cerebrovascular accidents 18 (1.0) 20 (0.5) .05
Major bleeding 102 (5.6) 114 (3.0) <.001
Blood transfusion (1 U) 244 (13.3) 99 (2.6) <.001
No. of unitsc <.001
1 53 (22.1) 7 (7.1)
2 102 (42.5) 33 (33.7)
>2 85 (35.4) 58 (59.2)
Acute kidney injuryd 76 (4.2) 41 (1.1) <.001
a
Data are presented as No. (percentage) of patients.
b
Anemia was defined as a hemoglobin level of <13 g/dL in men and <12 g/dL in women (to convert values to g/L, multiply by 10.0). No
anemia was defined as a hemoglobin level of 13 g/dL in men and 12 g/dL in women.
c
Four patients with and 1 patient without anemia had missing data for the number of units transfused (percentages are based on 240 and
98 patients, respectively).
d
Defined as documentation of “acute renal failure or acute kidney injury” as a complication in the medical record.

was used for in-hospital outcomes and Cox
proportional hazards models for long-term
outcomes. The covariates used in the models
were age (2 df spline), sex, diabetes, chronic
renal disease, body mass index (2 df spline),
hypertension, hyperlipidemia, smoking status,
history of MI, previous coronary artery bypass
graft, prior PCI, anticoagulation use before
PCI, history of cancer, history of cerebrovascu-
lar accident or transient ischemic attack,
peripheral vascular disease, congestive heart
failure status (present/history/never), ejection
fraction, NSTEMI, STEMI, estimated glomer-
ular filtration rate, left main coronary artery
PCI, left anterior descending artery PCI, left
circumflex artery PCI, right coronary artery
PCI, drug-eluting stent use, number of stents
placed, and number of diseased vessels. Statis-
tical significance is defined as a 2-tailed P
value of less than .05. Statistical analyses
were completed utilizing SAS statistical soft-
ware, version 9.4 (SAS Institute).
admission and 3837 (67.7%) did not. Patients
with anemia were older (mean age, 72.5 vs
64.8 years; P<.001) with a higher percentage
of women (35.8% [656] vs 27.2% [1042];
P<.001) than patients without anemia. Patients
with anemia had a higher prevalence of medical
comorbidities including diabetes mellitus,
chronic kidney disease, chronic heart failure, hy-
pertension, and hyperlipidemia, in addition to
previous MI, stroke, and percutaneous or surgi-
cal revascularization (Table 1). Patients with ane-
mia had a higher frequency of acute heart failure
(25.9% [474] vs 7.6% [291]; P<.001) and left
ventricular dysfunction (16.2% [297] vs 8.3%
317]; P<.001) on presentation compared to pa-
tients without anemia. The Sachdev index for co-
morbidity was higher in patients with anemia
(4.2 vs 2.0; P<.001). Malignancy, defined by
any reported history of solid tumor, lymphoma,
leukemia, or metastatic cancer, was 2-fold higher
in the anemic group (20.3% [371] vs 9.9%
[378]; P<.001).

RESULTS Clinical Presentation and Angiographic

Characteristics
Baseline Characteristics The most common ACS presentation was
Among 5668 patients who underwent PCI for NSTEMI in the 1831 patients with anemia
ACS between January 1, 2004, and December compared with the 3837 patients without ane-
31, 2014, 1831 (32.3%) had anemia on mia (45.3% [830] vs 30.9% [1187],
n n
1452 Mayo Clin Proc. October 2018;93(10):1448-1461 https://doi.org/10.1016/j.mayocp.2018.03.030
www.mayoclinicproceedings.org
BASELINE ANEMIA AND PCI OUTCOMES

P value
All-cause mortality Unadjusted 2.41 (1.76-3.31) <.001
Adjusted 1.04 (0.70-1.55) .85
Cardiac death Unadjusted 2.31 (1.64-3.25) <.001
Adjusted 1.02 (0.66-1.56) .93
Noncardiac death Unadjusted 2.95 (1.31-6.65) .009
Adjusted 1.16 (0.41-3.28) .78
MACE Unadjusted 1.79 (1.37-2.33) .001
Adjusted 0.96 (0.70-1.31) .78
Major bleeding Unadjusted 1.93 (1.47-2.53) .001
Adjusted 1.35 (1.00-1.83) .05
Recurrent infarction Unadjusted 0.99 (0.72-1.36) .94
Adjusted 0.85 (0.61-1.18) .33
Recurrent revascularization Unadjusted 0.72 (0.42-1.26) .24
Adjusted 0.73 (0.42-1.27) .26
Cerebrovascular accidents Unadjusted 1.89 (1.00-3.59) .05
Adjusted 0.98 (0.45-2.13) .96
Blood transfusion Unadjusted 5.81 (4.56-7.39) <.001
Adjusted 2.65 (2.02-3.48) <.001
Renal injury Unadjusted 4.01 (2.73-5.88) <.001
Adjusted 1.04 (0.61-1.77) .89

0.25 0.5 1.0 2.0 4.0 8.0

Odds ratio

FIGURE 1. In-hospital logistic regression for primary and secondary outcomes in patients with and without anemia before and after
adjustment for patient risk factors and comorbidities using inverse propensity weighting. MACE ¼ major adverse cardiac event.

respectively; P<.001), whereas patients
without anemia presented more commonly
with unstable angina (40.6% [1557] vs
32.1% [588]; P<.001) and STEMI (28.5%
[1093] vs 22.6% [413]; P<.001). Presentation
with reduced left ventricular ejection fraction
of 40% or less was 2-fold more common in
patients with anemia (P<.001). Moreover,
more patients with anemia had stage 3 or
worse chronic kidney disease on presentation
(P<.001) (Table 1). Left main coronary artery
disease (7.8% [141 of 1795] vs 2.6% [99 of
3754]; P<.001) and 2-vessel disease (16.4%
[294 of 1795] vs 13.2% [496 of 3754];
P¼.003) were more prevalent in patients
with anemia compared with patients without
anemia (Table 2).
Management
Radial access was less utilized in patients with
anemia compared with patients without
anemia (16.4% [294 of 1791] vs 20.2% [757
of 3745]; P<.001). Bare metal stents were
more commonly used in patients with anemia
than in those without anemia (27% [484] vs
17% [636]; P<.001). Ticagrelor and glycopro-
tein IIb/IIIa inhibitors were less commonly
used in patients with anemia compared with
patients without anemia (3.0% [55] vs 5.1%
[194] and 51.3% [940] vs 64.7% [2481],
respectively; both P<.001). There were no
other significant differences in guideline-
driven medical therapy (Table 2).
Crude Clinical Outcomes
In-Hospital Outcomes. The 1831 patients
with anemia had more than 2-fold higher
crude all-cause mortality than the 3837 pa-
tients without anemia (4.6% [84] vs 2.0%
[75]; P<.001), driven by increased cardiac
(3.8% [70] vs 1.7% [65]; P<.001) and
noncardiac (0.8% [14] vs 0.3% [10]; P¼.006)

Mayo Clin Proc. n October 2018;93(10):1448-1461 n https://doi.org/10.1016/j.mayocp.2018.03.030 1453

www.mayoclinicproceedings.org
 MAYO CLINIC PROCEEDINGS

TABLE 4. In-Hospital Outcomes of Patients With and Without Anemia After Performing Propensity Weighting
Analysisa
Variable Anemiab (n¼1831) No anemiab (n¼3837) P value
Primary outcomes
All-cause mortality 37 (2.0) 75 (2.0) .85
Cardiac death 32 (1.7) 65 (1.7) .93
Noncardiac death 6 (0.3) 10 (0.3) .78
Major adverse cardiovascular events 57 (3.1) 125 (3.3) .78
Secondary outcomes
Recurrent revascularization 17 (0.9) 49 (1.3) .26
Recurrent infarction 50 (2.7) 123 (3.2) .33
Cerebrovascular accidents 9 (0.5) 20 (0.5) .96
Major bleeding 73 (4.0) 114 (3.0) .05
Blood transfusion (1 U) 120 (6.6) 99 (2.6) <.001
No. of units transfusedc <.001
1 24 (20.3) 7 (7.1)
2 58 (49.3) 33 (33.7)
>2 36 (30.5) 58 (59.2)
Acute kidney injuryd 20 (1.1) 41 (1.1) .89
a
Data are presented as weighted frequencies (percentage) of patients. Weighted frequencies are rounded to the nearest integer.
b
Anemia was defined as a hemoglobin level of <13 g/dL in men and <12 g/dL in women (to convert values to g/L, multiply by 10.0). No
anemia was defined as a hemoglobin level of 13 g/dL in men and 12 g/dL in women.
c
Two patients with and 1 patient without anemia received less than 1 U of blood (percentages are based on 118 and 98 patients,
respectively).
d
Defined as documentation of “acute renal failure or acute kidney injury” as a complication in the medical record.

death. In-hospital bleeding events were more
frequent in patients with anemia compared
with patients without anemia (5.6% [102] vs
3.0% [114]; P<.001), as was the need for
blood transfusion (13.3% [244] vs 2.6% [99];
P<.001). In-hospital acute kidney injury was
also significantly more common in patients
with anemia (P<.001) (Table 3 and Figure 1).
Postdischarge Long-term Outcomes. Patients
were followed up for a median of 39 months
(25th, 75th percentiles, 12, 61 months), and
832 deaths were recorded. Anemia was associ-
ated with higher unadjusted 5-year all-cause
mortality (44.4% [509] in anemic vs 15.4%
[323] in nonanemic patients; P<.001)
(Supplemental Figure 1A, available online at
http://www.mayoclinicproceedings.org). The
percentage of crude long-term cardiac death
(14.4% [185] vs 4.9% [126]; P<.001) and
noncardiac death (27.1% [249] vs 9.2% [148];
P<.001) estimated by Kaplan-Meier method
was higher in patients with anemia at 5 years
post-PCI. Unadjusted MACE at 5 years post-
PCI was significantly higher in patients with
anemia compared with patients without
anemia (62.3% [769] vs 40.5% [1009];
P<.001) (Supplemental Figure 1B). The crude
n n
1454 Mayo Clin Proc.
incidence of recurrent MI and cerebrovascular
accidents was also significantly higher in pa-
tients with anemia compared with no anemia
(P<.001 and P¼.002, respectively)
(Supplemental Table 3, available online at
http://www.mayoclinicproceedings.org).
Outcomes in Balanced Cohort
In-Hospital Outcomes. After balancing for
confounders, the 1831 patients with anemia
and the 3837 without anemia had similar all-
cause (2.0% [37] vs 2.0% [75]; P¼.85), car-
diac, and noncardiac mortality (Supplemental
Table 4, available online at http://www.
mayoclinicproceedings.org), MACE (3.1%
[57] vs 3.3% [125]; P¼.78), and rates of
ischemic events (recurrent infarction, revas-
cularization, and cerebrovascular accidents)
(Table 4 and Figure 1). Major bleeding (4.0%
[73] vs 3.0% [114]; P¼.05) and blood trans-
fusion (6.6% [120] vs 2.6% [99]; P<.001)
rates, however, remained significantly higher
in patients with anemia even after IPW. Acute
kidney injury events were also similar between
groups at 1.1% (20) and 1.1% (41) (P¼.89).
Postdischarge Long-term Outcomes. After
applying IPW analysis, there was a trend of higher
October 2018;93(10):1448-1461 https://doi.org/10.1016/j.mayocp.2018.03.030
www.mayoclinicproceedings.org
BASELINE ANEMIA AND PCI OUTCOMES

40 20
Anemia Anemia
Nonanemia Nonanemia

Cardiac death (%)

30
Death (%)

P=.05 P=.63
20 10

10

0 0
0 1 2 3 4 5 0 1 2 3 4 5
Years from PCI Years from PCI
No. at risk No. at risk
Nonanemia 3762 2747 2347 1924 1423 891 Nonanemia 3762 2747 2347 1924 1423 891
Anemia 1747 1169 944 747 526 314 Anemia 1747 1169 944 747 526 314
A B

20 60
Anemia Anemia
50
Noncardiac death (%)

Nonanemia Nonanemia P=.32

40

MACE (%)
P=.04
10 30
20
10
0 0
0 1 2 3 4 5 0 1 2 3 4 5
No. at risk
Years from PCI No. at risk
Years from PCI
Nonanemia 3762 2747 2347 1924 1423 891 Nonanemia 3762 2747 2347 1924 1423 891
Anemia 1747 1169 944 747 526 314 Anemia 1747 1169 944 747 526 314
C D

FIGURE 2. Comparison of the risk of (A) all-cause death, (B) cardiac death, (C) noncardiac death, and (D) major adverse cardiac
events (MACE) over time after hospital discharge among patients with anemia vs patients without anemia, adjusted for patient risk
factors and comorbidities. PCI ¼ percutaneous coronary intervention.

all-cause 5-year mortality in patients with anemia
that did not reach statistical significance (17.8%
[203] vs 15.4% [323]; P¼.05; Figure 2A). Cardiac
mortality was not statistically different (5.8%
[71] vs 5.8% [126]; P¼.63; Figure 2B);
However, noncardiac mortality was significantly
higher in patients with anemia than without
anemia (10.2% [113] vs 7.1% [148]; P¼.04;
Figure 2C, and Supplemental Table 5, available
online at http://www.mayoclinicproceedings.
org). There were no significant differences in
MACE rates (37% [378] vs 40.5% [758]; P¼.32;
Figure 2D).
Moderate to Severe Anemia Outcomes
Among the 5668 patients, 748 (13.2%) had
moderate to severe anemia. Their outcomes
were compared with the 3837 patients without
anemia after balancing the risk profiles of the 2
groups using IPW analysis. No significant differ-
ences were seen in in-hospital mortality (1.7%
[13] vs 2.0% [75]; P¼.68) and MACE rates
(2.5% [19] vs 3.3% [125]; P¼.26) between the
2 groups. However, more in-hospital noncardiac
deaths (0.9% [6] vs 0.3% [10]; P¼.03) and trend
toward lower in-hospital cardiac mortality (0.9%
[6] vs 1.7% [65]; P¼.07) were observed in pa-
tients with moderate to severe anemia (Table 5,
Figure 3). Patients with moderate to severe ane-
mia had greater than 2-fold and greater than 4-
fold higher rates of major bleeding and blood
transfusion, respectively (P<.001 for both).
Moderate to severe anemia was associated with
lower recurrent revascularization rates
compared with patients without anemia (0.2%
[1] vs 1.3% [49]; P¼.001); No other differences

Mayo Clin Proc. n October 2018;93(10):1448-1461 n https://doi.org/10.1016/j.mayocp.2018.03.030 1455

www.mayoclinicproceedings.org
 MAYO CLINIC PROCEEDINGS

TABLE 5. In-Hospital Outcomes of Patients With Moderate to Severe Anemia vs No Anemia After Propensity
Weighting Analysisa
Moderate to severe
Variable anemiab (n¼748) No anemiab (n¼3837) P value
Primary outcomes
All-cause mortality 13 (1.7) 75 (2.0) .68
Cardiac death 6 (0.9%) 65 (1.7) .07
Noncardiac death 6 (0.9%) 10 (0.3) .03
Major adverse cardiovascular events 19 (2.5) 125 (3.3) .26
Secondary outcomes
Recurrent revascularization 1 (0.2) 49 (1.3) .001
Recurrent infarction 26 (3.5) 123 (3.2) .66
Cerebrovascular accidents 7 (0.9) 20 (0.5) .25
Major bleeding 45 (6.1) 114 (3.0) <.001
Blood transfusion 83 (11.1) 99 (2.6) <.001
No. of units transfused (1 U)c .03
1 15 (18.5) 7 (7.1)
2 27 (34.2) 33 (33.7)
>2 38 (47.3) 58 (59.2)
Acute kidney injuryd 13 (1.7) 41 (1.1) .17
a
Data are presented as weighted frequencies (percentage) of patients. Weighted frequencies are rounded to the nearest integer.
b
Moderate to severe anemia was defined as a hemoglobin level of <11 g/dL in men and women (to convert values to g/L, multiply by
10.0). No anemia was defined as a hemoglobin level of 13 g/dL in men and 12 g/dL in women.
c
Three patients with and 1 patient without anemia received less than 1 U of blood (percentages are based on 80 and 98 patients,
respectively).
d
Defined as documentation of “acute renal failure or acute kidney injury” as a complication in the medical record.

were observed in other ischemic events
(Table 5).
There was a significantly higher 5-year
mortality rate (22.3% [100 patients with ane-
mia] vs 15.4% [323 patients without anemia];
P<.001), driven by more noncardiac deaths
(10.4% [48] vs 7.1% [148]; P¼.006). MACE
were not significantly different in the 2 groups
at 5 years (34.6% [139] vs 40.5% [758];
P¼.08).
Secondary Analysis by Multivariate
Regression
As a secondary analysis, multivariate regression
was applied to the estimated risk-adjusted
effects of anemia on outcomes. The results
were similar to the IPW adjustment, although
the IPW approach generally resulted in adjusted
effects closer to the null (Supplemental Table 6
and Supplemental Figure 2, available online at
http://www.mayoclinicproceedings.org).
DISCUSSION
The present study provides one of the longest
follow-up periods for mortality and MACE in pa-
tients with ACS and anemia managed in the cur-
rent era of PCI and provides unique insights to
n n
1456 Mayo Clin Proc.
in-hospital and long-term causes of death.
Following adjustment of baseline differences us-
ing IPW, anemia at the time of presentation with
ACS in patients who undergo PCI is an indepen-
dent predictor for in-hospital major bleeding
and need for blood transfusion and is associated
with higher long-term noncardiac mortality.
Prevalence and Outcomes of Patients With
Anemia
Our analysis showed that 1 in 3 patients treated
with PCI for ACS had anemia, which is consis-
tent with prior registries5-11 but higher than
observed in patients enrolled in RCTs.3,4,12-14
Similar to prior studies,5,35-37 anemic patients
at our center had multiple comorbid conditions
that have been reported to be independent pre-
dictors of poor long-term survival, including
older age, diabetes mellitus, peripheral arterial
disease, chronic kidney disease, chronic heart
failure, left main coronary artery disease, previ-
ous MI, and prior cardiac interventions.28,38,39
Over the past decades, there was large interest
in determining if anemia has independent ef-
fects on cardiovascular outcomes of ACS. In
prior studies, multiple logistic regression ana-
lyses were used to account for differences in
October 2018;93(10):1448-1461 https://doi.org/10.1016/j.mayocp.2018.03.030
www.mayoclinicproceedings.org
BASELINE ANEMIA AND PCI OUTCOMES

P value
All-cause mortality Unadjusted 2.98 (2.03-4.39) <.001
Adjusted 0.88 (0.49-1.60) .68
Cardiac death Unadjusted 2.76 (1.81-4.22) <.001
Adjusted 0.51 (0.22-1.14) .07
Noncardiac death Unadjusted 4.14 (1.63-10.52) .005
Adjusted 3.34 (1.24-9.00) .03
MACE Unadjusted 2.13 (1.52-2.98) .001
Adjusted 0.76 (0.46-1.24) .26
Major bleeding Unadjusted 2.44 (1.74-3.42) <.001
Adjusted 2.11 (1.48-3.00) <.001
Recurrent infarction Unadjusted 1.09 (0.71-1.67) .71
Adjusted 1.10 (0.72-1.69) .66
Recurrent revascularization Unadjusted 0.84 (0.39-1.77) .63
Adjusted 0.12 (0.02-0.76) .001
Cerebrovascular accidents Unadjusted 1.54 (0.62-3.86) .37
Adjusted 1.72 (0.71-4.15) .29
Blood transfusion Unadjusted 11.79 (9.08-15.31) <.001
Adjusted 4.73 (3.49-6.40) <.001
Renal injury Unadjusted 5.92 (3.85-9.11) <.001
Adjusted 1.59 (0.84-3.00) .17

0.625 0.25 1.0 4.0 16.0

Odds ratio

FIGURE 3. In-hospital logistic regression for primary and secondary outcomes in patients with moderate to severe anemia and
without anemia before and after adjustment for patient risk factors and comorbidities using inverse propensity weighting. MACE ¼
major adverse cardiac event.

comorbid conditions and revealed persistence
of association of anemia with adverse cardiovas-
cular events, risk of bleeding, and death.3,5-14
However, other studies found that the increased
risk of mortality associated with anemia dissi-
pated after controlling for such con-
founders.4,26 A recent meta-analysis of 27 ACS
studies by Lewis et al25 involving 233,144 pa-
tients, 44,519 of whom were anemic, revealed
an increased risk of all-cause mortality (risk ra-
tio [RR], 2.08; 95% CI, 1.70-2.55) and reinfarc-
tion (RR, 1.25; 95% CI, 1.02-1.53) in the
anemic population over a median follow-up of
18 months; the risk for morality weakened
but continued to be higher after multivariate
analysis (RR, 1.49; 95% CI, 1.23-1.81). Similar
trends of adjusted mortality were reported in
another meta-analysis by Liu et al,40 who sug-
gested that underutilization of aspirin, b-
blocker, and statin therapy (73.8%, 41.4%,
and 49%, respectively) in patients with anemia
was one potential cause of increased risk of
death in the anemic population. Mamas et al10
recently reported the outcomes of anemic pa-
tients in a large UK Myocardial Ischemia Na-
tional Audit Project (MINAP) registry
including 422,855 patients diagnosed with
ACS. In their study, anemia was associated
with increased 30-day (odds ratio [OR], 1.28;
95% CI, 1.22-1.35) and 1-year (OR, 1.31;
95% CI, 1.27-1.35) adjusted all-cause mortal-
ity. Decreased evidence-based therapy utiliza-
tion was also noted. In contrast, at our center,
predismissal guideline-directed medical ther-
apy was highly utilized in both patients with
and without anemia, eliminating the potential
contribution of differences in management stra-
tegies on outcomes.
Anemia and Cause of Death
In this study with the longest follow-up to
date, we determined the cause of death in

Mayo Clin Proc. n October 2018;93(10):1448-1461 n https://doi.org/10.1016/j.mayocp.2018.03.030 1457

www.mayoclinicproceedings.org

patients presenting with anemia and
compared it with data in patients without ane-
mia. Importantly, using propensity adjust-
ment, we made the 2 groups similar, thereby
minimizing confounding of our results, and
studied anemia as an exposure variable. We
did not observe any differences in in-hospital
or long-term MACE or ischemic events be-
tween patients presenting with and without
anemia. The trend in higher all-cause mortality
was largely driven by noncardiac death. Ane-
mia was found previously by Kunadian
et al36 in their analysis of the ACUITY (Acute
Catheterization and Urgent Intervention
Triage Strategy Trial) study to independently
increase the absolute risk of cardiovascular
death and MI by 1.5% and 2.1% 1 year after
presenting with ACS. In the secondary analysis
of 16 TIMI (Thrombolysis in Myocardial
Infarction) trials, Sabatine et al4 found no as-
sociation between baseline Hbg level and risk
for 30-day cardiovascular death, MI, or
repeated intervention unless the Hgb level
decreased below 10 g/dL (adjusted OR, 2.69;
95% CI, 2.01-3.60; P<.001).
Cause of death has not been well studied
in patients with anemia who present with
ACS and undergo PCI. At our institution,
cause of death of all patients who have under-
gone PCI is rigorously ascertained and adjudi-
cated. In our study, the presence of anemia,
especially in moderate to severe range, was
highly associated with noncardiac causes of
death. Including older patients with higher
rates of comorbidity such as malignancy and
chronic renal insufficiency are potential rea-
sons for the observed differences between
our study findings and those derived from sec-
ondary analyses of RCTs. It is now increas-
ingly recognized that anemia may be a
marker of frailty.41 The higher prevalence of
anemia in patients with multiple comorbidities
is consistent with this observation. We did not
determine frailty in our study population but
can explain the higher long-term prevalence
of noncardiac deaths in patients with anemia.
The observed results in this study and the
lack of observed benefit by correcting anemia
in prior investigations argues for anemia as a
marker of higher medical comorbidity and
“poor survival” and may represent poor
cellular homeostasis. In addition, the comor-
bidity index was higher in patients with
n n
1458 Mayo Clin Proc. October 2018;93(10):1448-1461
MAYO CLINIC PROCEEDINGS
baseline anemia. Both of these factors worsen
the prognosis of patients undergoing PCI,
and anemia may be a marker of poor health.38
We previously reported higher mortality and
MI rates in frail patients who had undergone
PCI as compared with nonfrail patients,
underscoring the complex relationship be-
tween anemia, frailty, and adverse outcomes.42
Bleeding and Blood Transfusions in Patients
With Anemia
Our analysis supports the previously observed
increased risk of major bleeding and the
higher need for blood transfusion seen in
ACS patients with anemia even after balancing
baseline risk profiles. The absolute risk of in-
hospital major bleeding was 1.0% and 3.1%
higher in patients with anemia and moderate
to severe anemia, respectively, when
compared with controls without anemia. Rao
et al21,43 reported worse 30-day and
6-month all-cause mortality in patients with
severe bleeding after PCI (hazard ratio for 6-
month mortality, 7.5; 95% CI, 6.1-9.3;
P<.001). Moreover, in the ACUITY trial, ane-
mia was one of the predictors of major
bleeding, which was associated with a 3.5-
fold increase in 1-year all-cause mortality.44
In our analysis, blood transfusion rates were
2.5-fold and 4.5-fold in ACS patients with
anemia and moderate to severe anemia,
respectively, compared with nonanemic con-
trols. Lack of information regarding the timing
of blood transfusion or the Hgb level at the
time of blood transfusion limits our interpreta-
tion of the association between adverse cardio-
vascular events and need for blood transfusion
in the present study. In their pilot trial of con-
servative vs liberal red blood cell transfusion in
acute MI, Cooper et al15 found that liberal
blood transfusion was associated with a 25%
cumulative absolute risk increase in in-
hospital death, MI, and heart failure
(P¼.046). Subsequent meta-analysis of 10
studies reporting blood transfusion pattern in
ACS patients reported worse all-cause mortal-
ity (RR, 2.91; 95% CI, 2.46-3.44; P<.001) and
MI (RR, 2.04; 95% CI, 1.06-3.93; P¼.03) in
patients treated with liberal blood transfu-
sion.45 These observations are consistent
with the guidelines set by the American Asso-
ciation of Blood Banks recommending restric-
tive blood transfusion strategy in critically ill
https://doi.org/10.1016/j.mayocp.2018.03.030
www.mayoclinicproceedings.org
BASELINE ANEMIA AND PCI OUTCOMES
patients.46,47 Moreover, the American College
of Physicians guideline for treatment of anemia
in patients with heart disease suggests a trans-
fusion Hgb threshold of 7 to 8 g/dL in patients
hospitalized with coronary artery disease.48
Similarly, the American College of Cardiol-
ogy/American Heart Association advocates for
avoidance of blood transfusion in the absence
of ischemia in ACS patients unless the Hgb
level is 8 g/dL or lower,1 whereas the Euro-
pean Society of Cardiology guideline recom-
mends blood transfusion for those with
anemia in the presence of any of the following:
active bleeding, compromised hemodynamic
status, Hgb level less than 7 g/dL, or hemato-
crit level less than 25%.49
Study Limitations
The data on anemia and its severity and in-
hospital and follow-up outcomes were derived
from a single-center registry, and our findings
may not be generalizable to other patient sub-
sets (ACS managed conservatively, stable
angina, or use of coronary artery bypass graft-
ing). There were important baseline con-
founders between patients with and without
anemia that were accounted for using the
IPW method. Our registry included almost
all the variables that have been previously
linked to adverse clinical outcomes in ACS pa-
tients. However, other confounders that were
not measured may have influenced the find-
ings in this patient population. We obtained
postdismissal information on hospitalization
(reasons and outcomes), repeated revasculari-
zation, and adverse outcomes on all patients.
However, information on adherence (or non-
adherence) to guideline-directed medical ther-
apy and status of anemia postdischarge may
not have been consistently captured in the
database. Malignancy is an important
confounder in the data set and was not classi-
fied according to the current American Cancer
Society guidelines but was balanced for using
the propensity score method.
CONCLUSION
After accounting for differences in risk profiles
of anemic and nonanemic patients, anemia
appeared to be an independent risk factor for
increased long-term all-cause and noncardiac
mortality. It also increased the risk for in-
hospital major bleeding and blood transfusion.
Mayo Clin Proc. n October 2018;93(10):1448-1461
www.mayoclinicproceedings.org
SUPPLEMENTAL ONLINE MATERIAL
Supplemental material can be found online at
http://www.mayoclinicproceedings.org. Sup-
plemental material attached to journal articles
has not been edited, and the authors take re-
sponsibility for the accuracy of all data.
Abbreviations and Acronyms: ACS = acute coronary
syndrome; Hgb = hemoglobin; IPW = inverse propensity
weighting; MACE = major adverse cardiac events; MI =
myocardial infarction; NSTEMI = noneST-elevation MI; OR
= odds ratio; PCI = percutaneous coronary intervention;
RCT = randomized controlled trial; RR = risk ratio; STEMI =
ST-elevation MI
Potential Competing Interests: The authors report no
competing interests.
Data Previously Presented: These data were presented in
part at the 2016 American College of Cardiology meeting in
Chicago, IL.
Correspondence: Address to Mandeep Singh, MD, MPH,
Division of Cardiovascular Diseases, Mayo Clinic, 200 First
St SW, Rochester, MN 55905 (singh.mandeep@mayo.edu).
REFERENCES
1. Amsterdam EA, Wenger NK, Brindis RG, et al. 2014 AHA/ACC
guideline for the management of patients with e acute coronary
syndromes: a report of the American College of Cardiology/
American Heart Association task force on practice guidelines
[published correction appears in J Am Coll Cardiol. 2014;64(24):
2713e2714]. J Am Coll Cardiol. 2014;64(24):e139-e228.
2. O’Gara PT, Kushner FG, Ascheim DD, et al. 2013 ACCF/AHA
guideline for the management of ST-elevation myocardial
infarction: a report of the American College of Cardiology
Foundation/American Heart Association Task Force on Prac-
tice Guidelines. J Am Coll Cardiol. 2013;61(4):e78-e140.
3. Anker SD, Voors A, Okonko D, et al; OPTIMAAL Investigators.
Prevalence, incidence, and prognostic value of anaemia in pa-
tients after an acute myocardial infarction: data from the Opti-
maal trial. Eur Heart J. 2009;30(11):1331-1339.
4. Sabatine MS, Morrow DA, Giugliano RP, et al. Association of
hemoglobin levels with clinical outcomes in acute coronary
syndromes. Circulation. 2005;111(16):2042-2049.
5. Hasin T, Sorkin A, Markiewicz W, Hammerman H, Aronson D.
Prevalence and prognostic significance of transient, persistent,
and new-onset anemia after acute myocardial infarction. Am J
Cardiol. 2009;104(4):486-491.
6. Kurek T, Lenarczyk R, Kowalczyk J, et al. Effect of anemia in
high-risk groups of patients with acute myocardial infarction
treated with percutaneous coronary intervention. Am J Cardiol.
2010;105(5):611-618.
7. Al Falluji N, Lawrence-Nelson J, Kostis JB, Lacy CR, Ranjan R.
Wilson AC; Myocardial Infarction Data Acquisition system
(MIDAS #8) Study Group. Effect of anemia on 1-year mortality
in patients with acute myocardial infarction. Am Heart J. 2002;
144(4):636-641.
8. Ang DS, Kao MP, Noman A, Lang CC, Struthers AD. The prog-
nostic significance of early and late anaemia in acute coronary
syndrome. QJM. 2012;105(5):445-454.
9. Peterson ED, Dai D, DeLong ER, et al; NCDR Registry Partic-
ipants. Contemporary mortality risk prediction for percuta-
neous coronary intervention: results from 588,398
n https://doi.org/10.1016/j.mayocp.2018.03.030 1459

procedures in the National Cardiovascular Data Registry. J Am
Coll Cardiol. 2010;55(18):1923-1932.
10. Mamas MA, Kwok CS, Kontopantelis E, et al. Relationship be-
tween anemia and mortality outcomes in a national acute cor-
onary syndrome cohort: insights from the UK Myocardial
Ischemia National Audit Project registry. J Am Heart Assoc.
2016;5(11):e003348.
11. Lipsic E, van der Horst IC, Voors AA, et al. Hemoglobin levels
and 30-day mortality in patients after myocardial infarction
[published correction appears in Int J Cardiol. 2007;123(1):74].
Int J Cardiol. 2005;100(2):289-292.
12. Tsujita K, Nikolsky E, Lansky AJ, et al. Impact of anemia on clin-
ical outcomes of patients with ST-segment elevation myocar-
dial infarction in relation to gender and adjunctive
antithrombotic therapy (from the HORIZONS-AMI trial). Am
J Cardiol. 2010;105(10):1385-1394.
13. Nikolsky E, Aymong ED, Halkin A, et al. Impact of anemia in
patients with acute myocardial infarction undergoing primary
percutaneous coronary intervention: analysis from the
Controlled Abciximab and Device Investigation to Lower
Late Angioplasty Complications (CADILLAC) trial. J Am Coll
Cardiol. 2004;44(3):547-553.
14. Giraldez RR, Sabatine MS, Morrow DA, et al. Baseline hemo-
globin concentration and creatinine clearance composite labo-
ratory index improves risk stratification in ST-elevation
myocardial infarction. Am Heart J. 2009;157(3):517-524.
15. Cooper HA, Rao SV, Greenberg MD, et al. Conservative versus
liberal red cell transfusion in acute myocardial infarction (the
CRIT Randomized Pilot Study). Am J Cardiol. 2011;108(8):
1108-1111.
16. Levy PS, Quigley RL, Gould SA. Acute dilutional anemia and
critical left anterior descending coronary artery stenosis impairs
end organ oxygen delivery. J Trauma. 1996;41(3):416-423.
17. Willis P, Voeltz MD. Anemia, hemorrhage, and transfusion in
percutaneous coronary intervention, acute coronary syn-
dromes, and ST-segment elevation myocardial infarction. Am J
Cardiol. 2009;104(5, suppl):34C-38C.
18. Moscucci M, Fox KA, Cannon CP, et al. Predictors of major bleeding
in acute coronary syndromes: the Global Registry of Acute Coro-
nary Events (GRACE). Eur Heart J. 2003;24(20):1815-1823.
19. Manoukian SV, Feit F, Mehran R, et al. Impact of major bleeding
on 30-day mortality and clinical outcomes in patients with
acute coronary syndromes: an analysis from the ACUITY Trial.
J Am Coll Cardiol. 2007;49(12):1362-1368.
20. Nikolsky E, Mehran R, Sadeghi HM, et al. Prognostic impact of
blood transfusion after primary angioplasty for acute myocardial
infarction: analysis from the CADILLAC (Controlled Abciximab
and Device Investigation to Lower Late Angioplasty Complica-
tions) trial. J Am Coll Cardiol Intv. 2009;2(7):624-632.
21. Rao SV, Jollis JG, Harrington RA, et al. Relationship of blood
transfusion and clinical outcomes in patients with acute coro-
nary syndromes. JAMA. 2004;292(13):1555-1562.
22. Voeltz MD, Patel AD, Feit F, Fazel R, Lincoff AM,
Manoukian SV. Effect of anemia on hemorrhagic complications
and mortality following percutaneous coronary intervention.
Am J Cardiol. 2007;99(11):1513-1517.
23. Mehran R, Pocock S, Nikolsky E, et al. Impact of bleeding on mor-
tality after percutaneous coronary intervention: results from a
patient-level pooled analysis of the REPLACE-2 (Randomized
Evaluation of PCI Linking Angiomax to Reduced Clinical Events),
ACUITY (Acute Catheterization and Urgent Intervention Triage
Strategy), and HORIZONS-AMI (Harmonizing Outcomes With
Revascularization and Stents in Acute Myocardial Infarction) trials.
J Am Coll Cardiol Intv. 2011;4(6):654-664.
24. McKechnie RS, Smith D, Montoye C, et al; Blue Cross Blue Shield
of Michigan Cardiovascular Consortium (BMC2). Prognostic impli-
cation of anemia on in-hospital outcomes after percutaneous cor-
onary intervention. Circulation. 2004;110(3):271-277.
25. Lawler PR, Filion KB, Dourian T, Atallah R, Garfinkle M,
Eisenberg MJ. Anemia and mortality in acute coronary
n n
1460 Mayo Clin Proc. October 2018;93(10):1448-1461
MAYO CLINIC PROCEEDINGS
syndromes: a systematic review and meta-analysis. Am Heart
J. 2013;165(2):143-153.e5.
26. Valeur N, Nielsen OW, McMurray JJ, Torp-Pedersen C,
Køber L; TRACE Study Group. Anaemia is an independent pre-
dictor of mortality in patients with left ventricular systolic
dysfunction following acute myocardial infarction. Eur J Heart
Fail. 2006;8(6):577-584.
27. Singh M, Lennon RJ, Holmes DR Jr, Bell MR, Rihal CS. Corre-
lates of procedural complications and a simple integer risk score
for percutaneous coronary intervention. J Am Coll Cardiol. 2002;
40(3):387-393.
28. Singh M, Rihal CS, Lennon RJ, Spertus J, Rumsfeld JS,
Holmes DR Jr. Bedside estimation of risk from percutaneous
coronary intervention: the new Mayo Clinic risk scores [pub-
lished correction appears in Mayo Clin Proc. 2013;88(5):532].
Mayo Clin Proc. 2007;82(6):701-708.
29. Nutritional anaemias: report of a WHO scientific group. World
Health Organ Tech Rep Ser. 1968;405:5-37.
30. Sachdev M, Sun JL, Tsiatis AA, Nelson CL, Mark DB, Jollis JG.
The prognostic importance of comorbidity for mortality in pa-
tients with stable coronary artery disease. J Am Coll Cardiol.
2004;43(4):576-582.
31. Spoon DB, Psaltis PJ, Singh M, et al. Trends in cause of death
after percutaneous coronary intervention. Circulation. 2014;
129(12):1286-1294.
32. Raphael CE, Spoon DB, Bell MR, et al. Effect of preprocedural
thrombocytopenia on prognosis after percutaneous coronary
intervention. Mayo Clin Proc. 2016;91(8):1035-1044.
33. Austin PC. An introduction to propensity score methods for
reducing the effects of confounding in observational studies.
Multivariate Behav Res. 2011;46(3):399-424.
34. D’Agostino RB Jr. Propensity score methods for bias reduction
in the comparison of a treatment to a non-randomized control
group. Stat Med. 1998;17(19):2265-2281.
35. Wu WC, Rathore SS, Wang Y, Radford MJ, Krumholz HM.
Blood transfusion in elderly patients with acute myocardial
infarction. N Engl J Med. 2001;345(17):1230-1236.
36. Kunadian V, Mehran R, Lincoff AM, et al. Effect of anemia on
frequency of short- and long-term clinical events in acute cor-
onary syndromes (from the Acute Catheterization and Urgent
Intervention Triage Strategy Trial). Am J Cardiol. 2014;114(12):
1823-1829.
37. Alexander KP, Chen AY, Wang TY, et al; CRUSADE Investiga-
tors. Transfusion practice and outcomes in non-ST-segment
elevation acute coronary syndromes. Am Heart J. 2008;
155(6):1047-1053.
38. Singh M, Rihal CS, Roger VL, et al. Comorbid conditions and
outcomes after percutaneous coronary intervention. Heart.
2008;94(11):1424-1428.
39. Antman EM, Cohen M, Bernink PJ, et al. The TIMI risk score for un-
stable angina/non-ST elevation MI: a method for prognostication
and therapeutic decision making. JAMA. 2000;284(7):835-842.
40. Liu Y, Yang YM, Zhu J, Tan HQ, Liang Y, Li JD. Anaemia and
prognosis in acute coronary syndromes: a systematic review
and meta-analysis. J Int Med Res. 2012;40(1):43-55.
41. Afilalo J, Lauck S, Kim DH, et al. Frailty in older adults undergo-
ing aortic valve replacement: the FRAILTY-AVR study. J Am Coll
Cardiol. 2017;70(6):689-700.
42. Singh M, Rihal CS, Lennon RJ, Spertus JA, Nair KS, Roger VL.
Influence of frailty and health status on outcomes in patients
with coronary disease undergoing percutaneous revasculariza-
tion. Circ Cardiovasc Qual Outcomes. 2011;4(5):496-502.
43. Rao SV, O’Grady K, Pieper KS, et al. A comparison of the clin-
ical impact of bleeding measured by two different classifications
among patients with acute coronary syndromes. J Am Coll Car-
diol. 2006;47(4):809-816.
44. Mehran R, Rao SV, Bhatt DL, et al. Standardized bleeding def-
initions for cardiovascular clinical trials: a consensus report from
the Bleeding Academic Research Consortium. Circulation. 2011;
123(23):2736-2747.
https://doi.org/10.1016/j.mayocp.2018.03.030
www.mayoclinicproceedings.org
BASELINE ANEMIA AND PCI OUTCOMES
45. Chatterjee S, Wetterslev J, Sharma A, Lichstein E, Mukherjee D.
Association of blood transfusion with increased mortality in
myocardial infarction: a meta-analysis and diversity-adjusted
study sequential analysis. JAMA Intern Med. 2013;173(2):132-
139.
46. Carson JL, Grossman BJ, Kleinman S, et al; Clinical Transfusion
Medicine Committee of the AABB. Red blood cell transfusion:
a clinical practice guideline from the AABB. Ann Intern Med.
2012;157(1):49-58.
47. Carson JL, Stanworth SJ, Roubinian N, et al. Transfusion
thresholds and other strategies for guiding allogeneic red
blood cell transfusion. Cochrane Database Syst Rev. 2016;10:
CD002042.
Mayo Clin Proc. n October 2018;93(10):1448-1461 n https://doi.org/10.1016/j.mayocp.2018.03.030
www.mayoclinicproceedings.org
48. Qaseem A, Humphrey LL, Fitterman N, Starkey M, Shekelle P;
Clinical Guidelines Committee of the American College of Phy-
sicians. Treatment of anemia in patients with heart disease: a
clinical practice guideline from the American College of Physi-
cians [published correction appears in Ann Intern Med. 2014;
160(2):144]. Ann Intern Med. 2013;159(11):770-779.
49. Roffi M, Patrono C, Collet JP, et al; ESC Scientific Document
Group. 2015 ESC guidelines for the management of acute cor-
onary syndromes in patients presenting without persistent ST-
segment elevation: Task Force for the Management of Acute
Coronary Syndromes in Patients Presenting Without Persistent
ST-Segment Elevation of the European Society of Cardiology
(ESC). Eur Heart J. 2016;37(3):267-315.
1461