Objectives Background Methods: A B C D

Download as pdf or txt
Download as pdf or txt
You are on page 1of 10

Correction

Some abbreviations were confusing in the published article [Am Heart J 2010;160:1130-1136.e3]. These have been
expanded for clarity in this updated version.

Use of intensive lipid-lowering therapy in patients


hospitalized with acute coronary syndrome: An
analysis of 65,396 hospitalizations from 344 hospitals
participating in Get With The Guidelines (GWTG)
Usman Javed, MD, a Prakash C. Deedwania, MD, a Deepak L. Bhatt, MD, MPH, b Christopher P. Cannon, MD, b
David Dai, PhD, c Adrian F. Hernandez, MD, MHS, c Eric D. Peterson, MD, MPH, c and Gregg C. Fonarow, MD d
Fresno, and Los Angeles, CA; Boston, MA; and Durham, NC

Objectives The study aimed to analyze the use of intensive lipid-lowering therapy (I-LLT) at discharge in a broad
population of patients hospitalized with acute coronary syndrome (ACS).
Background Early and intensive statin therapy in ACS has been shown to reduce cardiovascular morbidity and
mortality. Utilization and predictors of I-LLT among hospitalized ACS patients are not known.
Methods The GWTG database was analyzed for ACS-related hospitalizations from 2005 to 2009. The use of I-LLT
(defined as dose of statin or combination therapy likely to produce N50% reductions in low-density lipoprotein [LDL]) and less
intensive lipid-lowering therapy (LI-LLT) at discharge was assessed. Baseline characteristics and temporal trends in LLT were
compared in these 2 treatment groups.
Results Of 65,396 patients receiving LLT, only 25,036 (38.3%) were treated with an I-LLT regimen. Mean total cholesterol,
LDL, and triglycerides were significantly higher in the I-LLT group. Even among those with LDL N130 mg/dL, 50% or less received
I-LLT. Predictors of I-LLT at discharge included LLT before admission, hyperlipidemia, prior coronary artery disease, increasing
body mass index, and in-hospital percutaneous coronary intervention. Although there was some temporal improvement in the
rate of I-LLT from 2005 to 2007, a decline in use of I-LLT was noted in 2008 and 2009. This was attributed to a sharp reduction
in use of ezetimibe in combination with statin, without corresponding increases in intensive statin monotherapy.
Conclusions In a large cohort of patients admitted with ACS, most of the eligible patients were not discharged on I-LLT.
These data suggest the need for better implementation of guideline-recommended intensive statin therapy in patients with ACS.
(Am Heart J 2011;161:418-424.e3.)

Several large studies have consistently demonstrated CAD and acute coronary syndrome (ACS), statin therapy
that lipid-lowering therapy (LLT) with 3-hydroxy-3- has shown a reduction in mortality and recurrent cardiac
methylglutaryl coenzyme A reductase inhibitors (statins) events.2-6 These data have established a very early clinical
reduce cardiovascular risk irrespective of underlying benefit that persisted on long-term follow-up. The PROVE
coronary artery disease (CAD).1 In patients with stable IT-TIMI 225 and MIRACL6 trials have shown even better
clinical outcomes with early and intensive statin therapy
From the aUniversity of California, San Francisco-Fresno Medical Education Program, in ACS. It is also well established that the adherence to the
Fresno, CA, bVA Boston Healthcare System and Brigham and Women's Hospital, Harvard use of statin therapy in the post-ACS patient is directly
Medical School, Boston, MA, cDuke University Medical Center, Durham, NC, and related to statin initiation during the index admission.7
d
University of California, Los Angeles School of Medicine, Los Angeles, CA.
Vera Bittner, MD, MSPH served as guest editor for this article.
In light of above the evidence, the recent National
Reprint requests: Prakash C. Deedwania, MD, Cardiology Section, UCSF Program at Cholesterol Education Program Adult Treatment Panel
Fresno, 2615 E. Clinton Avenue (111), Fresno, CA, 93703. guideline update recommended an optional low-density
E-mail: deed1@sbcglobal.net
lipoprotein (LDL) treatment goal of b70 mg/dL for
0002-8703/$ - see front matter
© 2011, Mosby, Inc. All rights reserved. patients with ACS.8,9 Moreover, the current guidelines
doi:10.1016/j.ahj.2010.12.014 of the American College of Cardiology/American Heart
American Heart Journal
Volume 161, Number 2
Javed et al 419

Association (ACC/AHA) recommend measurement of statin of any dose used in combination with ezetimibe (statin/
lipid levels on admission and instituting LLT before ezetimibe). All other LLTs were considered as less intensive LLT
hospital discharge in patients with ACS.10,11 (LI-LLT). A secondary analysis excluding ezetimibe and statin
The objective of our study was to assess the use of combination was performed to assess use and trends in intensive
statin monotherapy.
intensive lipid lowering therapy (I-LLT) at time of
Data collected included patient demographics, pertinent
discharge in patients admitted with ACS along with medical history, symptoms on arrival, laboratory results, inhos-
patient and hospital characteristics associated with use of pital treatment and procedures, discharge treatment, risk factor
I-LLT. This study analyzed data from the hospitals counseling, and patient disposition. The lipid levels obtained
participating in AHA's GWTG-CAD program from 2005 within the first 24 hours of hospitalization were measured at the
to 2009. Temporal trends in use of I-LLT were also local hospital laboratory. Yearly trends in admission lipid values
assessed. In patients admitted with ACS, prescription of and I-LLT were assessed from 2005 to 2009.
various LLTs (various agents and their prescribed doses)
at time of the hospital discharge was also assessed in Statistical analysis
relation to the patients' admission lipid profile with the Patients were divided into the I-LLT and LI-LLT categories as
probability of achieving LDL goal of b100 mg/dL and LDL defined above. In addition, I-LLT rate were noted in various
b70 mg/dL. subgroups based on admission LDL and high-density lipoprotein
(HDL). In the descriptive analysis, the mean (±SD) and
percentages were reported for continuous and categorical
Methods variables, respectively. For comparison of baseline characte-
GWTG-CAD is a national initiative of the AHA to promote ristics in I-LLT and LI-LLT groups, Wilcoxon rank-sum tests
guidelines adherence in management of hospitalized patients were used for continuous variables and χ2 tests for categorical
with coronary artery disease. The data collection process used in variables. In examining the association between LDL and I-LLT,
this study and quality control features have been previously a multivariable logistic regression was used. The generalized
described.12 All participating institutions were granted waiver of estimating equation (GEE) method with exchangeable working
informed consent by their local institutional review boards. The correlation structure was used to account for within-hospital
Duke Clinical Research Institute (Durham, NC) serves as the data clustering because patients at the same hospital are more
analysis center and has an agreement to analyze the aggregate likely to have similar responses relative to patients in other
de-identified data for research purposes. The GTWG program is hospitals (ie, within-center correlation for response). The
supported by the American Heart Association in part through an method produces estimates similar to those from ordinary
unrestricted education grant from the Merck Schering Plough logistic regression, but the estimated variances of the estimates
Partnership that did not participate in the design, analysis, are adjusted for the correlation of outcomes within each
preparation, review, or approval of the manuscript. The authors hospital. The variables entered into the model are patient
are solely responsible for the design and conduct of this study, age, gender, race, body mass index, cardiovascular risk factors
all study analyses, the drafting and editing of the paper and its (smoking, hypertension, hyperlipidemia, diabetes mellitus, renal
final contents. insufficiency, prior MI stroke, heart failure, LLT before
admission), and type of ACS. A sensitivity analysis (28,724
Study population subjects at 76 sites), confined to the centers with N70% statin
This study was drawn from 159,713 admissions with the medication dose reporting compliance, was used to exclude
diagnosis of ACS (including ST-segment elevation myocardial any selection bias in the primary analysis. A P value of b.05
infarction [STEMI], non-STEMI [NSTEMI], and unstable angina), was considered significant for the test of each variable. All
between July 2005 and December 2009, from 410 participating analyses were performed using SAS software (version 9.2, SAS
hospitals across the United States. Patients were excluded if they Institute, Cary, NC) by the Duke Clinical Research Institute
left against medical advice, discontinued care, died, or were (Durham, NC).
discharged to a federal hospital, hospice, or another acute care
hospital. Of the 138,216 patients discharged, 119,387 (86.4%)
were receiving LLT and 14,279 (10.3%) were discharged Results
without LLT. Lipid-lowering therapy was contraindicated in The clinical characteristics of the patient study
4,550 (3.3%). Of patients discharged on LLT, 53,991 admissions population are shown in Table I. Admission diagnosis
were also excluded because the details describing agent/dose
was MI in 91.7% patients, while the remaining patients
was missing. The data from 65,396 admissions at 344 sites were
complete for the purposes of this analysis and formed the final
had unstable angina. There were 41.7% of patients who
study population. Appendix Table IA (online) shows the were receiving LLT before the index ACS admission.
characteristics of patients included and excluded from the Admission LDL levels were assessed in 54,892 (83.9%) of
study population. patients. The characteristics of patients with and without
LDL levels assessed are shown in Appendix Table IB
Lipid-lowering therapy (online). Patients without lipid testing during hospitali-
Intensive lipid-lowering therapy was defined as therapy likely zation were more likely to have been receiving LLT
to achieve a N50% reduction in LDL and included atorvastation before admission. At hospital discharge, there were
40 or 80 mg, rosuvastatin 20 or 40 mg, simvastatin 80 mg, or any 25,036 (38.3%) patients receiving I-LLT and 40,360
American Heart Journal
420 Javed et al February 2011

Table I. Baseline characteristics in intensive and less intensive LLT groups


Patient characteristics Overall (N = 65 396) Intensive LLT (n = 25 036) Less intensive LLT (n = 40 360) P value

Age (y) 64.7 ± 13.9 62.6 ± 13.4 66.0 ± 14.1 b.001


Female 34.3% 32.2% 35.6% b.001
Race/ethnicity
White 72.3% 71.4% 72.9% b.001
Black 7.0% 7.7% 6.5% b.001
Hispanic 6.4% 6.2% 6.5% .149
Asian 2.8% 2.9% 2.8% .220
Diagnosis
STEMI/non-STEMI 91.7% 92.4% 91.4% b.001
Unstable Angina 8.3% 7.6% 8.6% b.001
LLT taken before Admission 41.7% 44.4% 40.0% b.001
Prior myocardial infarction 19.9% 21.0% 19.2% b.001
Prior stroke 8.2% 7.5% 8.6% b.001
Peripheral vascular disease 8.1% 8.0% 8.2% .341
Hypertension 67.8% 67.5% 67.9% .319
Diabetes—IDDM 9.3% 9.9% 8.9% b.001
Diabetes—NIDDM 15.9% 15.9% 15.9% .993
Hyperlipidemia 55.7% 58.7% 53.7% b.001
Smoking (current or prior 1 y) 33.5% 35.8% 32.0% b.001
β-Blockers 97.7% 98.1% 97.4% b.001
ACE inhibitors 72.9% 75.5% 71.2% b.001
ARBs 12.5% 12.4% 12.5% .724
Aspirin 98.2% 98.6% 98.0% b.001
Clopidogrel 80.8% 84.9% 78.3% b.001
Warfarin 10.5% 10.2% 10.6% .187
Nitrates 27.0% 27.8% 26.6% b.001
Calcium channel blockers 9.0% 8.4% 9.4% b.001
Aldosterone blockers 3.6% 3.8% 3.5% .066
Total cholesterol (mg/dL) 170.1 ± 48.2 174.6 ± 51.0 167.1 ± 46.2 b.001
LDL cholesterol (mg/dL) 103.4 ± 40.0 107.2 ± 43.0 101.0 ± 37.7 b.001
HDL cholesterol (mg/dL) 38.1 ± 12.4 37.9 ± 11.9 38.3 ± 12.7 .254
Triglycerides (mg/dL) 155.4 ± 124.7 161.0 ± 128.1 151.8 ± 122.3 b.001

ARB, Angiotensin receptor blocker; ACE, angiotensin converting enzyme; IDDM, insulin dependent diabetes mellitus; NIDDM, non-insulin dependent diabetes mellitus.

(61.7%) receiving LI-LLT. Patients receiving I-LLT were Factors associated with I-LLT
younger, less likely to be female, and had higher A number of patient characteristics were more frequent
admission LDL levels (Table I). There were 30.0% of in patients discharged with I-LLT (Table I). Diagnosis of
patients who received statin monotherapy, whereas 8.2% STEMI, presence of ST changes/LBBB on admission ECG,
received statin/ezetimibe. The characteristics of patients and PCI with or without stent, elevated total cholesterol,
receiving statin monotherapy and those receiving statin/ LDL, and triglyceride values were more likely to be
ezetimibe combination are shown in Appendix Table II discharged on I-LLT. There was no impact of uninsured
(online). Among various statins, the rate of use and status, non–insulin-dependent diabetes mellitus, prior
dosage of various statins in I-LLT subgroup are shown in CABG, or HDL between the 2 groups. Table II provides
Appendix Table III (online). the rates of I-LLT based on admission LDL and HDL levels.
When the analysis was confined to 76 hospitals that Multivariate analysis of these data using the GEE model
collected statin dose in N70% of patients (n = 28,724), the demonstrated LLT before admission, history of CAD or
findings were similar (39.3% on I-LLT, of which 33.1% prior MI, hyperlipidemia, LDL per 10 mg/dL rise, body
were on statin monotherapy). In comparison, the mass index (BMI) increase by 5 units, PCI with stent
patients without LLT dose documentation had lower placement, and male gender as independent predictors
rates of prior LLT, diabetes, hyperlipidemia (lower total of I-LLT. Patients with increasing age, chronic dialysis,
choles-terol, LDL cholesterol, and triglycerides), estab- and unstable angina had a lower likelihood of receiving
lished CAD, prior CABG or percutaneous coronary I-LLT (Figure 1). There was a marginal impact of
intervention (PCI), and acute STEMI. These patients confining the analysis to intensive statin monotherapy
however had higher prevalence of hypertension, PVD, and center reporting compliance on GEE model results.
prior MI or cerebrovascular accident, and NSTEMI. The In the statin monotherapy model, diabetes mellitus,
excluded sites also had lower rates of revascularization prior PCI, and prior CABG were additional independent
(PCI or CABG) and teaching hospitals. predictors of I-LLT.
American Heart Journal
Volume 161, Number 2
Javed et al 421

Table II. Use of intensive LLT at discharge based on admission HDL-C and LDL-C levels
LDL (mg/dL)

100-130 130-160
HDL (mg/dL) <70 (n = 9 157) 70-100 (n = 13 603) (n = 11 918) (n = 6 672) ≥ 160 (n = 3 738) Total (n = 45 088)

<40 (n = 27 762) 13.29% 18.71% 16.15% 8.76% 4.66% 61.57%


38.19% 36.03% 38.68% 44.14% 52.38% 39.58%
40-60 (n = 14 589) 5.70% 9.55% 8.73% 5.28% 3.10% 32.36%
36.95% 35.53% 37.68% 43.32% 52.58% 39.26%
≥ 60 (n = 2 737) 1.31% 1.92% 1.56% 0.76% 0.53% 6.07%
33.45% 30.44% 34.33% 38.71% 44.96% 34.38%
Total (n = 45 088) 20.31% 30.17% 26.43% 14.80% 8.29% 100%
37.53% 35.51% 38.09% 43.57% 51.98% 39.16%

Frequency missing 20,308. Numbers in bold denote the percentage of ACS hospitalizations in that cell that received intensive LLT at hospital discharge.

2007). However, an insignificant drop in rate of I-LLT was


Figure 1 noted with a decline to 35.7% by December 2009. We
found this to be primarily due to a significant drop in the
use of statin/ezetimibe combination from 11.4% in 2007
to 3.4% in 2009 (Table III). When statin/ezetimibe
combination was excluded as I-LLT, less than one third
of ACS patients was treated with intensive statin
monotherapy. Use of intensive statin monotherapy at
discharge was 28.0% in 2005 and 33.1% in 2009, without
significant change during the 2007 to 2009 period.

Discussion
The present analysis shows that among hospitals
participating in GWTG-CAD, most hospitalized ACS
patients are not discharged on I-LLT. Even among those
with admission LDL N130 mg/dL, 50% or less received
I-LLT. During the first 3 years of observation in this study,
there was very modest improvement in I-LLT on discharge.
This trend did not persist, instead a decline in this
therapeutic approach was observed during 2008 to 2009.

Role of I-LLT in ACS


Although statins play a pivotal role in LDL reduction,
they may also exhibit a pleotropic effect by decreasing
extent of myocardial ischemia, remodeling, as well as
promoting plaque stabilization and endothelial func-
tion.13-15 Based on these mechanistic properties, and as
demonstrated in several clinical studies, it is now widely
accepted that initiation of an early and intensive statin
therapy in ACS is associated with reduced inpatient
mortality and morbidity3,4,6,16 as well as improved
longterm survival and lower rates of recurrent coronary
Factors associated with I-LLT by multivariate GEE model.
events. In the MIRACL trial, I-LLT with atorvastatin
80 mg/d (vs placebo) was started within 24 to 96 hours
of presentation with ACS.6 It was associated with a
Temporal trends in the use of I-LLT were also lower risk of symptomatic ischemia requiring emergent
examined. I-LLT rates since the publication of updated rehospitalization. This effect was independent of base-
National Cholesterol Education Program-ATP guidelines line LDL level, although LDL was decreased from 126 to
in 2004 increased initially from 35.5% to 41.6% (2005 to 72 mg/dL in the treatment group. The clinical benefit
American Heart Journal
422 Javed et al February 2011

Table III. Temporal trends of intensive LLT and intensive statin monotherapy
Intensive LLT overall Yearly Intensive statin Yearly Ezetimibe plus Yearly
Year Total (N = 65 396) (n = 25 036) trend P monotherapy (n = 19 645) trend P statin (n = 5391) trend P

2005 5283 1875 (35.49) 1422 (26.92) 453 (8.57)


2006 15 520 6108 (39.36) .039 4516 (29.10) .448 1592 (10.26) .043
2007 18 082 7523 (41.60) .138 5467 (30.23) .381 2056 (11.37) .202
2008 17 143 6188 (36.10) b.001 5220 (30.45) .204 968 (5.65) b.001
2009 9368 3342 (35.67) .221 3020 (32.24) .734 322 (3.44) .002

started to exhibit at 4 weeks and then persisted for the LDL N160 mg/dL were still left untreated with I-LLT. Thus,
duration of the study. The PROVE-IT TIMI 22 trial has these patients had a low probability of achieving target
further demonstrated that aggressive LLT in ACS, with LDL in near future.
even lower targets LDL levels, leads to reduction in It is interesting to note that in our study, I-LLT was more
revascularization and unstable angina.3,6 In PROVE-IT likely to be used in younger patients, male, smokers,
TIMI 22, the median LDL was decreased to 62 mg/dL on overweight patients, in those with STEMI and otherwise
80 mg of atorvastatin in comparison to 95 mg/dL on those more likely to undergo PCI, and those with high
40 mg of pravastatin. Similar to the findings in the lipid levels. Ironically, patients with diagnosis of unstable
MIRACL, the beneficial effects of high-dose statins angina, prior stroke, heart failure, and renal insufficiency
emerged as early as 30 days and then persisted during were treated with less LI-LLT. Moreover, there was no
the 2 years of follow-up. difference in the type of therapy in those with prior
Based on the available evidence, the revised Adult CABG and non–insulin-dependent diabetes mellitus. The
Treatment Panel III guidelines has recommended early present study demonstrates the underutilization of I-LLT
and I-LLT in patients admitted for ACS and has included in the very high risk group, which is prone to recurrent
an optional therapeutic goal of LDL b70 mg/dL in these ischemic cardiovascular events.
high-risk patients.8,9 Our analysis shows that despite
available evidence and recommendations, in this large I-LLT at discharge
cohort of hospitalized patient with ACS, 10.3% (n = The available evidence suggests better long-term
14,279) of patients with ACS were not discharged on LLT. compliance and higher survival rates in ACS patients
Moreover, only 38.2% of eligible patients were dis- initially discharged on statin therapy than those who
charged on I-LLT. Although LDL remains the primary were not. 7 Subsequently, the CRUSADE Quality
goal for therapeutic intervention, the I-LLT prescribed at Improvement Initiative also showed that the use of LLT
the time of discharge may also take HDL into consider- at discharge among select ACS patients increased from
ation. The inverse relationship of HDL and with nonfatal 78% in 2000 to 88% in 2004. 21 The overall low rate of
MI and cardiovascular-related death has been demon- I-LLT observed in our analyses along with recent declines
strated previously.17 The present study illustrates that during 2008 to 2009 are concerning and emphasize the
61.6% of patients presenting with ACS have HDL levels need for implementation of evidence-based and guide-
b40 mg/dL. To improve secondary prevention of line-recommended therapy in most patients with ACS.
cardiovascular risk, it may be necessary to implement Although intensive statin monotherapy continued to
additional lipid-modifying therapy (together with routine increase marginally, the drop-off in 2008 and 2009 in
statin therapy) targeting HDL N40 mg/dL in males and intensive LLT was essentially due to the decrease in use of
N50 mg/dL in females. ezetimibe in combination in statin therapy. However,
there was no offsetting increased use of intensive statin
Lipid measurement in hospitalized patients with ACS therapy. These data represent interesting, but potentially
Although the current guidelines recommend lipid unfortunate consequences of the well-publicized EN-
measurement in ACS, it is measured in less than half of HANCE trial controversy.22-25 In its aftermath, the
these patients in routine clinical practice.18 This practice plethora of discussion raised further controversy about
has been largely based on the convention that lipid levels even the proven benefits of statin therapy. This has had a
are unreliable in ACS settings and usually associated with large impact on the contemporary practice of lipid
an initial decrement in total cholesterol and LDL.19 management without providing a clear alternative to
However, more recent data have shown less pronounced the use of ezetimibe. As a result, starting in 2008, fewer
changes in lipid profile.20 In this analysis, about half of ACS patients were treated with therapy that would allow
ACS patients had LDL b100 mg/dL, with I-LLT used in them to achieve LDL cholesterol goals recommended in
about 36% of such patients. Although rate of I-LLT national guidelines. Despite the lower use of statin/
increased with the rise in LDL, nearly half of patients with ezetimibe combination, there was little to no shift to
American Heart Journal
Volume 161, Number 2
Javed et al 423

highdose statin therapy (at least at time of hospital Conclusions


discharge). These data underscore yet another impact on During the period of 2005 to 2009, only about one third
routine clinical practice heralded by safety or lack of of patients hospitalized with ACS were discharged on
efficacy concerns as raised by some recent controver- I-LLT. Even among patients with documented admission
sies.26,27 Nevertheless, ezetimibe alone or in combination LDL, which would require N50% reduction to achieve an
with statin therapy has not been proven to change optional goal of LDL b70 mg/dL, only about 50% were
outcomes in ACS, so the full clinical implications of these discharged on I-LLT. Independent predictors of I-LLT at
treatment shifts are not yet known. discharge included LLT before admission, history of
hyperlipidemia or coronary artery disease, increasing BMI
Limitations
and lipid level, and in-hospital percutaneous coronary
There is a potential for selection bias in this study
intervention. In addition, the rate of adopting I-LLT in ACS
because discharge LLT dosing data were not available in
decreased significantly over the last 2 years because of a
50% of patients. There were modest baseline differences
marked decline in the use of ezetimibe in combination
between those with and without discharge dosing of lipid
with statin therapy without an offsetting increase in
therapy recorded. These factors may influence the
intensive statin monotherapy. These findings underscore
generalizability of these findings. Furthermore, the
the importance of ongoing emphasis regarding imple-
GWTG-CAD database is voluntary and therefore may
mentation of current guidelines for measuring lipids and
not be representative of the entire US practice. These
intensive statin therapy in all ACS patients.
findings may not reflect care at hospitals that differ
substantially from participating hospitals. Registry hospi-
tals tend to be larger than nonparticipating hospitals, are
more likely to be affiliated with a medical school, and are Disclosures
more likely to have available facilities for cardiac
Dr Bhatt: research grants from Astra Zeneca, Bristol-
catheterization, PCI, and cardiac surgery. GWTG-CAD
Myers Squibb, Eisai, Ethicon, Heartscape, Sanofi Aventis
participating hospitals also were provided with feedback
and the Medicines Company. Dr Deedwania: consultant/
on performance that may have also influenced the care
advisory board of AstraZeneca and Pfizer. Dr Peterson:
patterns. The hospitals participating in the GWTG-CAD
research funding from Bristol Myers Squibb, Sanofi
program may be more likely to prescribe I-LLT, such that
Aventis partnership. Dr Cannon: research grants from
the treatment gaps are even larger than what was
Accumetrics, AstraZeneca, Bristol-Myers Squibb/Sanofi
observed here. Hence, the data presented here might
Partnership, Glaxo Smith Kline Intekrin Therapeutics,
reflect different and possibly higher rates of I-LLT than
Novartis, Takeda, clinical advisor and equity in Auto-
actually prescribed among patients and hospitals that
medics Medical Systems. Dr Hernandez: research grant
differ from those participating in GWTG-CAD. Although
from Johnson and Johnson, Merck, and honorarium from
the lipid levels obtained in this study were measured in
AstraZeneca and Medtronic. Dr Fonarow: consultant/
the first 24 hours of admission, they may or may not be
advisory board to Merck Schering Plough and honorar-
entirely reflective of the baseline steady-state lipid levels.
ium from Abbott, AstraZeneca, Merck Schering Plough,
Furthermore, we do not have data as to whether patients
and Pfizer. Other authors have no disclosures.
were in the fasting state. This real-world study used
results of various commercially available lipid panel
assays rather than results from a single central core
laboratory. Although this methodology may introduce References
great variability to lipid testing results, this approach
1. Baigent C, Keech A, Kearney PM, et al. Efficacy and safety of
makes these findings more applicable to clinical practice. cholesterol-lowering treatment: prospective meta-analysis of data
This study only assessed LLT dosing at time of hospital from 90,056 participants in 14 randomised trials of statins. Lancet
discharge. Some patients may have had subsequent 2005;366:1267-78.
change in dosing or modification of their lipid therapy 2. Blazing MA, De Lemos JA, Dyke CK, et al. The A-to-Z Trial: methods
regimen as an outpatient. As the current guidelines do not and rationale for a single trial investigating combined use of low-
specify a dose of statin for ACS patients but only a target molecular-weight heparin with the glycoprotein IIb/IIIa inhibitor
of therapy (ie, optional target of LDL b70 mg/dL in high tirofiban and defining the efficacy of early aggressive simvastatin
risk patients), many clinicians may believe that the therapy. Am Heart J 2001;142:211-7.
3. Cannon CP, McCabe CH, Belder R, et al. Design of the Pravastatin or
titration to I-LLT can occur post discharge, and this may
Atorvastatin Evaluation and Infection Therapy (PROVE IT)-TIMI 22
explain the reason for the treatment gap. The utilization trial. Am J Cardiol 2002;89:860-1.
of I-LLT at hospital discharge does not necessarily indicate 4. de Lemos JA, Blazing MA, Wiviott SD, et al. Early intensive vs a
that patients remained adherent to their discharge delayed conservative simvastatin strategy in patients with acute
regimen. The extent of dietary and exercise counseling coronary syndromes: phase Z of the A to Z trial. JAMA 2004;292:
were not available in this study. 1307-16.
American Heart Journal
424 Javed et al February 2011

5. Cannon CP, Braunwald E, McCabe CH, et al. Intensive versus 14. Every NR, Frederick PD, Robinson M, et al. A comparison of the
moderate lipid lowering with statins after acute coronary syndromes. national registry of myocardial infarction 2 with the cooperative
N Engl J Med 2004;350:1495-504. cardiovascular project. J Am Coll Cardiol 1999;33:1886-94.
6. Kinlay S, Schwartz GG, Olsson AG, et al. High-dose atorvastatin 15. Di Napoli P, Antonio Taccardi A, et al. Simvastatin reduces
enhances the decline in inflammatory markers in patients with acute reperfusion injury by modulating nitric oxide synthase expression: an
coronary syndromes in the MIRACL study. Circulation 2003;108: ex vivo study in isolated working rat hearts. Cardiovasc Res 2001;51:
1560-6. 283-93.
7. Muhlestein JB, Horne BD, Bair TL, et al. Usefulness of in-hospital 16. Fonarow GC, Wright RS, Spencer FA, et al. Effect of statin use
prescription of statin agents after angiographic diagnosis of coronary within the first 24 hours of admission for acute myocardial infarction
artery disease in improving continued compliance and reduced on early morbidity and mortality. Am J Cardiol 2005;96:611-6.
mortality. Am J Cardiol 2001;87:257-61. 17. Gordon T, Castelli WP, Hjortland MC, et al. High density lipoprotein
8. Third Report of the National Cholesterol Education Program (NCEP) as a protective factor against coronary heart disease: The Framing-
Expert Panel on Detection, Evaluation, and Treatment of High Blood ham Study. Am J Med 1977;62:707-14.
Cholesterol in Adults (Adult Treatment Panel III) Final Report. 18. Ko DT, Alter DA, Newman AM, et al. Association between lipid
Circulation 2002;106:3143-421. testing and statin therapy in acute myocardial infarction patients. Am
9. Grundy SM, Cleeman JI, Merz CNB, et al, for the Coordinating Heart J 2005;150:419-25.
Committee of the National Cholesterol Education Program and 19. Fyfe T, Baxter RH, Cochran KM, et al. Plasma-lipid changes after
Endorsed by the National Heart, Lung, and Blood Institute, American myocardial infarction. Lancet 1971;2:997-1001.
College of Cardiology Foundation, and American Heart Association. 20. Pitt B, Loscalzo J, Ycas J, et al. Lipid levels after acute coronary
Implications of Recent Clinical Trials for the National Cholesterol syndromes. J Am Coll Cardiol 2008;51:1440-5.
Education Program Adult Treatment Panel III Guidelines. Circulation 21. Mehta RH, Roe MT, Chen AY, et al. Recent trends in the care of
2004;110:227-39. patients with non-ST-segment elevation acute coronary syndromes:
10. Antman EM, Anbe DT, Armstrong PW, et al. ACC/AHA guidelines insights from the CRUSADE initiative. Arch Intern Med 2006;166:
for the management of patients with ST-elevation myocardial 2027-34.
infarction—executive summary. A report of the American College of 22. Simvastatin with or without ezetimibe in familial hypercholesterol-
Cardiology/American Heart Association Task Force on Practice emia. N Engl J Med 2008;358:1431-43.
Guidelines (writing committee to revise the 1999 guidelines for the 23. Carey J. Do cholesterol drugs do any good? Bus Week 2008:52-9.
management of patients with acute myocardial infarction). J Am Coll 24. Taubes G. What's cholesterol got to do with it? N Y Times 2008.
Cardiol 2004;44:671-719. 25. Greenland P, Lloyd-Jones D. Critical Lessons From the ENHANCE
11. Braunwald E, Antman EM, Beasley JW, et al. ACC/AHA 2002 Trial. JAMA 2008;299:953-5.
guideline update for the management of patients with unstable 26. Roe MT, Chen AY, Cannon CP, et al. Temporal changes in the use of
angina and non–ST-segment elevation myocardial infarction— drug-eluting stents for patients with non–ST-segment–elevation
summary article. A report of the American College of Cardiology/ myocardial infarction undergoing percutaneous coronary interven-
American Heart Association Task Force on Practice Guidelines tion from 2006 to 2008: results from the Can Rapid risk stratification
(Committee on the Management of Patients With Unstable Angina). of Unstable angina patients Suppress ADverse outcomes with Early
J Am Coll Cardiol 2002;40:1366-74. implementation of the ACC/AHA guidelines (CRUSADE) and Acute
12. Sachdeva A, Cannon CP, Deedwania PC, et al. Lipid levels in patients Coronary Treatment and Intervention Outcomes Network–Get With
hospitalized with coronary artery disease: an analysis of 136,905 The Guidelines (ACTION–GWTG) Registries. Circ Cardiovasc Qual
hospitalizations in Get With The Guidelines. Am Heart J 2009;157: Outcomes 2009;2:414-20.
111-7.e2. 27. Atwater BD, Oujiri J, Wolff MR. The immediate impact of the Clinical
13. Lefer AM, Campbell B, Shin YK, et al. Simvastatin preserves the Outcomes Utilizing Revascularization and Aggressive Drug Evalua-
ischemic-reperfused myocardium in normocholesterolemic rat hearts. tion (COURAGE) trial on the management of stable angina. Clin
Circulation 1999;100:178-84. Cardiol 2009;32:E1-3.
American Heart Journal
Volume 161, Number 2
Javed et al 424.e1

Appendix

Table IA. Patient characteristics of the study and excluded patients

Patient characteristics Overall (n = 159 713) Excluded (n = 94 317) Study cohort (N = 65 396) P value

Age (y) 66.2 ± 14.4 67.3 ± 14.6 64.7 ± 13.9 b.001


Female 35.6% 36.5% 34.3% b.001
Race/ethnicity
White 70.6% 69.4% 72.3% b.001
Black 7.4% 7.7% 7.0% b.001
Hispanic 5.6% 5.1% 6.4% b.001
Asian 3.2% 3.5% 2.8% b.001
Diagnosis
STEMI/non-STEMI 93.6% 94.9% 91.7% b.001
Unstable angina 6.4% 5.1% 8.3% b.001
LLT taken before admission
Prior myocardial infarction 20.1% 20.4% 19.9% .042
Prior stroke 8.9% 9.6% 8.2% b.001
Peripheral vascular disease 8.7% 9.2% 8.1% b.001
Hypertension 68.2% 68.6% 67.8% .003
Diabetes—IDDM 8.2% 7.1% 9.3% b.001
Diabetes—NIDDM 14.1% 12.3% 15.9% b.001
Hyperlipidemia 51.3% 47.2% 55.7% b.001
Smoking (current or prior 1 y) 31.1% 29.5% 33.5% b.001
β-Blockers 96.9% 96.3% 97.7% b.001
ACE inhibitors 70.8% 68.8% 72.9% b.001
ARBs 12.2% 12.1% 12.5% .043
Aspirin 97.4% 96.7% 98.2% b.001
Clopidogrel 75.6% 69.9% 80.8% b.001
Warfarin 11.3% 12.4% 10.5% b.001
Nitrates 17.8% 11.4% 27.0% b.001
Calcium channel blockers 6.5% 4.7% 9.0% b.001
Aldosterone blockers 3.7% 3.9% 3.6% .102
Total cholesterol (mg/dL) 168.7 ± 48.2 167.5 ± 48.1 170.1 ± 48.2 b.001
LDL cholesterol (mg/dL) 102.1 ± 40.0 101.1 ± 40.1 103.4 ± 40.0 b.001
HDL cholesterol (mg/dL) 38.5 ± 12.7 38.8 ± 13.1 38.1 ± 12.4 b.001
Triglycerides (mg/dL) 152.9 ± 122.3 150.6 ± 120.0 155.4 ± 124.7 b.001
PCI with stent 35.4% 25.1% 50.2% b.001
CABG 7.5% 6.5% 8.7% b.001

ARB, Angiotensin receptor blocker; ACE, angiotensin converting enzyme; IDDM, insulin dependent diabetes mellitus; NIDDM, non-insulin dependent diabetes mellitus.
American Heart Journal
424.e2 Javed et al February 2011

Table IB. Patient characteristics based on measurement of lipid levels

Patient characteristics Overall (N = 65 396) Lipids measured (n = 54 892) Lipids not measured (n = 10 504) P value

Age (y) 64.7 ± 13.9 64.0 ± 13.8 68.4 ± 13.9 b.001


Female 34.3% 33.6% 38.0% b.001
Race/ethnicity
White 72.3% 72.4% 71.9% .290
Black 7.0% 7.1% 6.3% .002
Hispanic 6.4% 6.4% 6.2% .427
Asian 2.8% 2.4% 5.0% b.001
Diagnosis
STEMI/non-STEMI 91.7% 91.9% 91.1% .008
Unstable angina 8.3% 8.1% 8.9% .008
LLT taken before Admission 41.7% 38.7% 57.2% b.001
Prior myocardial infarction 19.9% 18.9% 24.9% b.001
Prior stroke 8.2% 7.7% 10.9% b.001
Peripheral vascular disease 8.1% 7.6% 10.7% b.001
Hypertension 67.8% 67.1% 71.4% b.001
Diabetes—IDDM 9.3% 8.4% 13.7% b.001
Diabetes —NIDDM 15.9% 15.0% 20.8% b.001
Hyperlipidemia 55.7% 55.2% 58.2% b.001
Smoking (current or prior 1 y) 33.5% 34.9% 26.2% b.001
β-Blockers 97.7% 97.9% 96.3% b.001
ACE inhibitors 72.9% 74.0% 66.8% b.001
ARBs 12.5% 12.0% 15.1% b.001
Aspirin 98.2% 98.5% 97.0% b.001
Clopidogrel 80.8% 82.2% 73.5% b.001
Warfarin 10.5% 10.4% 11.0% .081
Nitrates 27.0% 26.8% 28.1% .008
Calcium channel blockers 9.0% 8.5% 11.8% b.001
Aldosterone blockers 3.6% 3.4% 4.7% b.001
PCI with Stent 50.2% 52.4% 38.9% b.001
CABG 8.7% 8.6% 9.0% .147
Intensive LLT 38.3% 38.9% 35.2% b.001
Intensive statin monotherapy 30.0% 30.8% 26.2% b.001
Statin/ezetimibe 8.2% 8.1% 8.9% .004

Table II. Patient characteristics in intensive LLT groups: overall, statin monotherapy, and ezetimibe plus any statin

Patient characteristics Overall I-LLT (n = 25 036) Statin monotherapy (n = 19 645) Statin/ezetimibe (n = 5 391) P value

Age (y) 62.6 ± 13.4 62.4 ± 13.5 63.6 ± 12.8 b.001


Female 32.2% 31.7% 34.0% .001
Race/ethnicity
White 71.4% 69.6% 77.9% b.001
Black 7.7% 8.1% 6.3% b.001
Hispanic 6.2% 6.8% 4.2% b.001
Asian 2.9% 3.1% 2.2% b.001
Diagnosis
STEMI/non-STEMI 92.4% 93.5% 88.3% b.001
Unstable angina 7.6% 6.5% 11.7% b.001
LLT taken before admission 44.4% 41.3% 56.0% b.001
Prior myocardial infarction 21.0% 19.8% 25.2% b.001
Prior stroke 7.5% 7.3% 8.2% .038
Peripheral vascular disease 8.0% 7.4% 10.0% b.001
Hypertension 67.5% 66.3% 71.7% b.001
Diabetes—IDDM 9.9% 9.6% 11.1% .002
Diabetes—NIDDM 15.9% 15.6% 16.8% .050
Hyperlipidemia 58.7% 55.5% 70.0% b.001
Smoking (current or prior 1 y) 35.8% 37.3% 30.5% b.001
Total cholesterol (mg/dL) 174.6 ± 51.0 175.6 ± 49.9 170.8 ± 54.5 b.001
LDL cholesterol (mg/dL) 107.2 ± 43.0 108.5 ± 42.5 102.1 ± 44.6 b.001
HDL cholesterol (mg/dL) 37.9 ± 11.9 37.9 ± 11.9 38.1 ± 11.9 .085
Triglycerides (mg/dL) 161.0 ± 128.1 158.8 ± 125.0 169.4 ± 138.4 b.001
American Heart Journal
Volume 161, Number 2
Javed et al 424.e3

Table III. Specific lipid-lowering agents in treatment groups

Overall
Statin Stain dose (N = 65 396) Overall (%) I-LLT (n = 25 036) I-LLT (%) LI-LLT (n = 40 360) LI-LLT (%)

Rosuvastatin 5 mg 649 0.99 31 0.12 618 1.53


10 mg 2116 3.24 85 0.34 2031 5.03
20 mg 1156 1.77 1156 4.62 0 0.00
40 mg 417 0.64 417 1.67 0 0.00
Atovastatin 10 mg 4122 6.30 52 0.21 4070 10.08
20 mg 5683 8.69 109 0.44 5574 13.81
40 mg 7331 11.21 7331 29.28 0 0.00
80 mg 7919 12.11 7919 31.63 0 0.00
Simvastatin 5 mg 94 0.14 2 0.01 92 0.23
10 mg 1086 1.66 24 0.10 1062 2.63
20 mg 6665 10.19 121 0.48 6544 16.21
40 mg 10 887 16.65 284 1.13 10 603 26.27
80 mg 3589 5.49 3589 14.34 0 0.00
Simvastatin/ezetimibe 10-10 mg 153 0.23 153 0.61 0 0.00
Combination
10-20 mg 1005 1.54 1005 4.01 0 0.00
10-40 mg 1815 2.78 1815 7.25 0 0.00
10-80 mg 603 0.92 603 2.41 0 0.00
Other 131 0.20 131 0.52 0 0.00
Statin/ezetimibe⁎ 1690 2.58 1690 6.75 0 0.00
⁎Ezetimibe and any dose of any statin in separate doses.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy