The n e w e ng l a n d j o u r na l of m e dic i n e

original article

Simvastatin for the Prevention

of Exacerbations in Moderate-to-Severe COPD
G.J. Criner, J.E. Connett, S.D. Aaron, R.K. Albert, W.C. Bailey, R. Casaburi,
J.A.D. Cooper, Jr., J.L. Curtis, M.T. Dransfield, M.K. Han, B. Make, N. Marchetti,
F.J. Martinez, D.E. Niewoehner, P.D. Scanlon, F.C. Sciurba, S.M. Scharf,
D.D. Sin, H. Voelker, G.R. Washko, P.G. Woodruff, and S.C. Lazarus, for the
COPD Clinical Research Network and the Canadian Institutes of Health Research

A BS T R AC T

BACKGROUND
Retrospective studies have shown that statins decrease the rate and severity of ex- The authors full names, degrees, and af-
acerbations, the rate of hospitalization, and mortality in chronic obstructive pul- filiations are listed in the Appendix. Ad-
dress reprint requests to Dr. Criner at 745
monary disease (COPD). We prospectively studied the efficacy of simvastatin in Parkinson Pavilion, Temple Lung Center,
preventing exacerbations in a large, multicenter, randomized trial. Temple University School of Medicine,
3401 N. Broad St., Philadelphia, PA, 19140,
METHODS or at [email protected].
We designed the Prospective Randomized Placebo-Controlled Trial of Simvastatin
in the Prevention of COPD Exacerbations (STATCOPE) as a randomized, controlled This article was published on May 18, 2014,
at NEJM.org.
trial of simvastatin (at a daily dose of 40 mg) versus placebo, with annual exacerba-
tion rates as the primary outcome. Patients were eligible if they were 40 to 80 years N Engl J Med 2014;370:2201-10.
of age, had COPD (defined by a forced expiratory volume in 1 second [FEV1] of less DOI: 10.1056/NEJMoa1403086
Copyright 2014 Massachusetts Medical Society.
than 80% and a ratio of FEV1 to forced vital capacity of less than 70%), and had a
smoking history of 10 or more pack-years, were receiving supplemental oxygen or
treatment with glucocorticoids or antibiotic agents, or had had an emergency de-
partment visit or hospitalization for COPD within the past year. Patients with dia-
betes or cardiovascular disease and those who were taking statins or who required
statins on the basis of Adult Treatment Panel III criteria were excluded. Participants
were treated from 12 to 36 months at 45 centers.
RESULTS
A total of 885 participants with COPD were enrolled for approximately 641 days;
44% of the patients were women. The patients had a mean (SD) age of 62.28.4
years, an FEV1 that was 41.617.7% of the predicted value, and a smoking history of
50.627.4 pack-years. At the time of study closeout, the low-density lipoprotein
cholesterol levels were lower in the simvastatin-treated patients than in those who
received placebo. The mean number of exacerbations per person-year was similar
in the simvastatin and placebo groups: 1.361.61 exacerbations and 1.391.73 ex-
acerbations, respectively (P=0.54). The median number of days to the first exacer-
bation was also similar: 223 days (95% confidence interval [CI], 195 to 275) and 231
days (95% CI, 193 to 303), respectively (P=0.34). The number of nonfatal serious
adverse events per person-year was similar, as well: 0.63 events with simvastatin
and 0.62 events with placebo. There were 30 deaths in the placebo group and 28 in
the simvastatin group (P=0.89).
CONCLUSIONS
Simvastatin at a daily dose of 40 mg did not affect exacerbation rates or the time to
a first exacerbation in patients with COPD who were at high risk for exacerbations.
(Funded by the National Heart, Lung, and Blood Institute and the Canadian Insti-
tutes of Health Research; STATCOPE ClinicalTrials.gov number, NCT01061671.)
n engl j med 370;23nejm.orgjune 5, 2014 2201
The New England Journal of Medicine
Downloaded from nejm.org on December 26, 2016. For personal use only. No other uses without permission.
Copyright 2014 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

C
hronic obstructive pulmonary dis- macy Current Good Manufacturing Practices Ser-
ease (COPD) is the third leading cause of vices. All the authors assume responsibility for
death in the United States.1 It is character- the data and analyses, the adherence of the study
ized by acute exacerbations that are associated to the protocol, and the completeness and accu-
with increased hospitalizations and costs of care, racy of the reported findings.
worsened quality of life, and increased mortality.2-9
Effective therapies for the treatment or prevention PARTICIPANT POPULATION
of exacerbations are limited. We enrolled patients, 40 to 80 years of age, with
Statins are 3-hydroxy-3-methylglutaryl co a clinical diagnosis of moderate-to-severe COPD,
enzyme A (HMG-CoA) reductase inhibitors that defined according to the Global Initiative for
reduce the risks of acute cardiac events and Chronic Obstructive Lung Disease (GOLD) criteria26
death.10-13 Although widely used for their lipid- (a ratio of forced expiratory volume in 1 second
lowering effects, statins are also reported to [FEV1] to forced vital capacity [FVC] of <70% after
have antiinflammatory effects.14-19 Multiple retro- bronchodilator use and an FEV1 of <80% of the
spective studies have shown that statins are ben- predicted value after bronchodilator use). Partici-
eficial in COPD. Reported benefits include re- pants were former or current smokers with a life-
ductions in rates of hospitalization (for COPD or time cigarette consumption of 10 or more pack-
any other reason), lung-function decline, the need years who also met at least one of the following
for mechanical ventilation, and death.20-24 How- criteria within the year before enrollment: use of
ever, except for one small, single-center, random- supplemental oxygen, receipt of systemic gluco-
ized study,25 all the studies that have shown ben- corticoids or antibiotic agents for respiratory prob-
eficial effects of statins in patients with COPD lems, or presentation to the emergency depart-
have been retrospective. The Prospective Random- ment or hospitalization for COPD exacerbation.
ized Placebo-Controlled Trial of Simvastatin in the We excluded patients who were already re-
Prevention of COPD Exacerbations (STATCOPE) ceiving statins; those who should have been re-
was a prospective, multicenter trial conducted by ceiving statins according to the Adult Treatment
the National Heart, Lung, and Blood Institute Panel III risk stratification27; those who were re-
(NHLBI) COPD Clinical Research Network to ex- ceiving drugs that are contraindicated with statins;
amine the effect of daily treatment with simva those who had active liver disease, alcoholism,
statin for at least 12 months (range, 12 to 36) on or allergy; and those who were unable to take
the rate of exacerbations among patients with statins. Patients who met the criteria for statin
moderate-to-severe COPD and no other indications treatment were excluded from the study on the
for statin treatment. basis of established guidelines.27 Before patients
were enrolled, lipid levels were measured after
ME THODS the patient had fasted for 12 hours.
After STATCOPE began on March 4, 2010, the
STUDY DESIGN AND OVERSIGHT Food and Drug Administration (FDA) issued a
In this randomized, parallel-group, placebo- warning (on June 8, 2011) against the concomi-
controlled trial, participants were randomly as- tant use of amlodipine or high doses of verapamil
signed in a 1:1 ratio to receive simvastatin orally with simvastatin.28 This announcement resulted
at a dose of 40 mg or an identical-appearing pla- in discontinuation of the study drug in partici-
cebo once daily. Participants were recruited from pants from both study groups who were receiv-
45 sites (29 sites in the United States and 16 in ing amlodipine or high-dose verapamil (16 pa-
Canada). Written informed consent was obtained tients in the simvastatin group and 20 in the
from all participants. The institutional review board placebo group). The STATCOPE data and safety
at each center approved the study protocol. The monitoring board subsequently recommended
complete protocol, including methods and the the exclusion of patients with diabetes, accord-
statistical analysis plan, is available with the full ing to the medical history or a glycated hemo-
text of this article at NEJM.org. The study drugs, globin level of more than 6.5%, which resulted in
simvastatin and placebo, were purchased, prepared, discontinuation of the study treatment in 28 par
and supplied by Temple University School of Phar- ticipants (14 in each group).

2202 n engl j med 370;23nejm.orgjune 5, 2014

The New England Journal of Medicine

Downloaded from nejm.org on December 26, 2016. For personal use only. No other uses without permission.
Copyright 2014 Massachusetts Medical Society. All rights reserved.
 Simvastatin for Prevention of Exacerbations in COPD

OUTCOMES ticipants were queried about the interim develop-

The primary outcome was the effect of simva ment of exacerbations and study-drug adherence.
statin on the exacerbation rate, which was defined
as the number of exacerbations per person-year. INTERIM ANALYSIS, STOPPING GUIDELINES,
Secondary outcomes included the time to the AND MONITORING PLAN
first exacerbation, the severity of exacerbations, The data and safety monitoring board met approx-
the number of acute cardiovascular events, qual- imately every 6 months to review recruitment and
ity of life, and changes in spirometric variables. follow-up rates, study-drug adherence, safety, and
Exacerbations were defined as an increase in efficacy results. Reviews of outcome data at the
severity or new onset of two or more respiratory meetings involved multiple statistical testing per-
symptoms (cough, sputum, wheezing, dyspnea, formed on a set of accumulating data, with the
or chest tightness) lasting for at least 3 days and use of a sequential monitoring plan that was
requiring treatment with antibiotics or systemic based on the alpha spending approach of Lan and
glucocorticoids.26 Exacerbation severity was graded DeMets.34
according to the following scale: mild (required The formal guidelines for stopping the study
only home management, with or without contact owing to futility were not part of the monitoring
with a health care provider), moderate (required a plan, but as specified in the protocol, the results
visit to an emergency department), severe (required of analyses that were based on stochastic cur-
hospitalization), and very severe (required intuba- tailment were presented to the data and safety
tion and mechanical ventilation). monitoring board. At the March 2013 meeting,
there were no indications of a treatment effect;
STUDY VISITS however, the board members voted to continue
Spirometric variables (FVC and FEV1) and disease- the study. At the October 2013 meeting, with
specific and general quality of life (assessed with 878 participants having undergone randomization,
the use of the St. Georges Respiratory Question- the primary outcome was the same in the two
naire [SGRQ]29 and the Medical Outcomes Study study groups (COPD exacerbation rate, 1.33 ex-
36-Item Short-Form Health Survey [SF-36],30 re- acerbations per person-year in each group). The
spectively) were measured at baseline and at 12, power of the study, which was conditional on these
24, and 36 months (and at the final visit if that observed rates, was 5%. There were no treatment
did not coincide with an annual visit). Scores on effects in any subgroup. The rates of moderate and
the SGRQ range from 0 to 100, with lower scores severe exacerbations did not differ significantly
indicating better functioning and with a mini- between the two study groups, and the numbers
mal clinically important difference of 4 points.29 of deaths were similar (27 deaths in the simva
The SF-36 is a self-administered general health statin group and 26 in the placebo group). The
questionnaire that generates scores across eight data and safety monitoring board therefore vot-
health dimensions. Scores range from 0 to 100, ed to stop STATCOPE for futility.
with higher scores indicating better health on
each dimension.30 Clinically important changes in STATISTICAL ANALYSIS
the SF-36 score are increases of 5 to 10 points in Sample-size and power considerations were based
specific domains.31 on the annual rate of exacerbations as the pri-
Spirometry was performed according to Amer- mary outcome. We estimated that the placebo
ican Thoracic SocietyEuropean Respiratory Soci- group would have an average of 1.54 exacerba-
ety guidelines.32 Data are presented as the percent tions per person-year and that simvastatin would
of reference predicted values, with the use of pre- reduce this rate by 15%. The estimated rate for
diction equations from Hankinson et al.33 Only the placebo group was based on the observed ex-
data obtained after the patient had used a broncho- acerbation rate in the placebo group of a previous
dilator were analyzed. trial of azithromycin conducted by the COPD
Throughout the trial, participants were fol- Clinical Research Network.8
lowed at clinic visits every 3 months and by means We based our sample-size calculations on a two-
of monthly telephone contacts during months sided alpha level of 0.05, a study power of 90%, the
without a clinic visit. During each contact, par- expected number of exacerbations per person-year

n engl j med 370;23nejm.orgjune 5, 2014 2203

The New England Journal of Medicine
Downloaded from nejm.org on December 26, 2016. For personal use only. No other uses without permission.
Copyright 2014 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

in the placebo group (1.54), the expected number R E SULT S

of exacerbations per person-year in the simva
statin group (1.540.85=1.31), a mean duration of ENROLLMENT AND FOLLOW-UP
follow-up of 2 years, and an equal probability Screening for the study, randomization, and follow-
of assignment to the placebo or simvastatin up of the participants are shown in Figure 1. Enroll-
group. We calculated that we would need to en- ment began March 4, 2010, and the last participant
roll 911 patients; assuming a 10% loss to follow-up, was enrolled on October 11, 2013. Participants were
the total sample-size estimate was 1002 patients. followed for a mean (SD) of 641354 days in the
An important secondary outcome was the time simvastatin group and 639351 days in the pla-
to the first exacerbation. Assuming that simva cebo group.
statin treatment would reduce the exacerbation
rate by 15% and adjusting for the 10% loss to CHARACTERISTICS OF THE PARTICIPANTS
follow-up, we estimated that 1126 patients would The demographic and baseline clinical charac-
be required in order to provide the study with teristics of the participants are shown in Table 1
90% power for this secondary outcome. Follow- (for complete demographic and baseline clinical
ing FDA recommendations and guidance from data, see Table S1 in the Supplementary Appen-
the data and safety monitoring board for the dix, available at NEJM.org). We randomly assigned
exclusion of patients who were receiving certain 885 participants with COPD to the two study groups.
drugs and the exclusion of patients with diabetes, A total of 44% of the patients were women. The
respectively, we projected that 74 additional par- patients had a mean age of 62.28.4 years, an
ticipants would be needed, for a final study popu- FEV1 that was 41.617.7% of the predicted value,
lation of 1200 patients. and a smoking history of 50.627.4 pack-years.
All the patients who underwent randomization There were no significant differences between the
were followed until the end of the study. The final two groups with respect to any characteristics at
analysis was performed on an intention-to-treat baseline. Follow-up visit information was not
basis. COPD exacerbation rates in the two study available for 8 of the 885 participants (0.9%), so
groups were compared with the use of a rate 430 patients in the simvastatin group and 447 in
ratio. The independence of individual exacerba- the placebo group were included in the primary
tions was ensured by considering participants to analysis.
have had two separate exacerbations if the onset
dates were at least 14 days apart. Exacerbation COPD EXACERBATIONS
rates in each group and the between-group dif- A total of 1982 acute COPD exacerbations oc-
ferences were analyzed with the use of negative curred during the study: 965 exacerbations among
binomial regression modeling and time-weighted the 430 patients in the simvastatin group and
intention-to-treat analyses with adjustments of con- 1017 exacerbations among the 447 patients in
fidence intervals for between-participant variation the placebo group. The COPD exacerbation rates
(overdispersion). were similar in the simvastatin and placebo
Secondary outcome measures were assessed groups: 1.361.61 exacerbations per person-year
at baseline and at 12, 24, and 36 months (and at and 1.391.73 exacerbations per person-year, re-
the final visit if it did not coincide with an annual spectively (P=0.54) (Fig. 2). We found no effect
visit). Analyses of the time to the first exacerba- of sex on the exacerbation rate.
tion and the rate of death from any cause were The median number of days to the first ex
performed with the use of KaplanMeier methods, acerbation was also similar in the two groups:
with survival curves for the probability of remain- 223 days (95% confidence interval [CI], 195 to
ing outcome-free in the two groups as a function 275) in the simvastatin group and 231 days
of time from randomization. Curves were com- (95%CI, 193 to 303) in the placebo group
pared with the use of the log-rank test. Continu- (P=0.34) (Fig. 3). The time to the first exacer-
ous outcome measures, including absolute and bation in the simvastatin group was slightly
relative changes in FEV1 and FVC, dyspnea assess- shorter for men than for women, but this dif-
ment, and SGRQ scores were computed as rates ference did not reach statistical significance
of change over time, from baseline to the date of (P=0.09) (Fig. S1 in the Supplementary Appendix).
the last observation. The severity of the exacerbations was not af-

2204 n engl j med 370;23nejm.orgjune 5, 2014

The New England Journal of Medicine

Downloaded from nejm.org on December 26, 2016. For personal use only. No other uses without permission.
Copyright 2014 Massachusetts Medical Society. All rights reserved.
 Simvastatin for Prevention of Exacerbations in COPD

1217 Patients were assessed for eligibility

332 Were excluded

97 Did not meet spirometric criteria
7 Had diabetes
21 Had liver disease
19 Were taking a contraindicated
medication
11 Were already taking a statin
38 Should have been taking a statin,
according to ATP-III criteria
5 Were unable to take statins
134 Had other reason

885 Underwent randomization

433 Were assigned to receive simvastatin 452 Were assigned to receive placebo

3 Were lost to follow-up 5 Were lost to follow-up

430 Were included in primary analysis 447 Were included in primary analysis

20 Withdrew
16 Withdrew
6 Lost interest
7 Lost interest
5 Were not willing to follow
1 Had difficulty accessing
protocol
clinic
3 Had difficulty accessing
5 Had medical condition
clinic
1 Did not want to be
1 Moved out of area
in control group
3 Had medical condition
1 Had adverse event
1 Had adverse event
1 Had other reason
1 Had other reason

316 (73.5%) Completed at least 12 mo 329 (73.6%) Completed at least 12 mo

189 (44.0%) Completed at least 24 mo 203 (45.4%) Completed at least 24 mo
86 (20.0%) Completed 36 mo 81 (18.1%) Completed 36 mo
28 (6.5%) Died during course of study 30 (6.7%) Died during course of study

Figure 1. Screening, Randomization, and Study Completion.

ATP-III denotes Adult Treatment Panel III.

fected by the study-drug assignment or sex. The simvastatin group; the difference was not sig-
number and severity of exacerbations in the sim- nificant). The study-drug assignment had no ef-
vastatin and placebo groups are shown in Tables S2 fect on the frequency of exacerbations (Fig. S2 in
and S3 in the Supplementary Appendix, and rates the Supplementary Appendix). There was no ef-
of exacerbation according to severity are shown in fect of study-drug assignment on the percentage
Table S4 in the Supplementary Appendix. of patients who received antibiotic or glucocorti-
A total of 296 of 877 participants (33.8%) coid therapy for an exacerbation (Table S8 in the
had three or more exacerbations during the study Supplementary Appendix). Smoking status, study
(155 patients in the placebo group and 141 in the center, patients age, severity of airflow obstruc-

n engl j med 370;23nejm.orgjune 5, 2014 2205

The New England Journal of Medicine
Downloaded from nejm.org on December 26, 2016. For personal use only. No other uses without permission.
Copyright 2014 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

tion (i.e., GOLD stage), and use or nonuse of oxy- and SGRQ, respectively (Table S5 in the Supple-
gen did not affect the outcomes (Fig. S3 in the mentary Appendix).
Supplementary Appendix).
ADVERSE EVENTS
SECONDARY OUTCOMES The most common adverse events, aside from ex-
There was no effect of simvastatin on lung func- acerbations, were pneumonia and other respira-
tion as assessed spirometrically. Likewise, simva tory and cardiovascular events. The rates of all
statin had no effect on the general or respiratory- nonfatal adverse events (the number of serious
specific quality of life, as assessed by the SF-36 adverse events per person-year) were similar in

Table 1. Characteristics of the Patients.*

Simvastatin Placebo
Characteristic (N=433) (N=452)
Age yr 62.28.5 62.38.4
Female sex no. (%) 184 (42.5) 203 (44.9)
FEV1 after bronchodilator use liters (% of predicted 1.20.6 (41.517.8) 1.20.6 (41.617.6)
value)
FEV1:FVC % 44.412.6 44.313.5
GOLD stage no./total no. (%)
2 142/430 (33.0) 145/449 (32.3)
3 142/430 (33.0) 163/449 (36.3)
4 146/430 (34.0) 141/449 (31.4)
Smoking history pack-yr 50.026.1 51.228.7
Current smoking no. (%) 133 (30.7) 143 (31.6)
COPD medication no. (%)
Inhaled glucocorticoids only 14 (3.2) 12 (2.7)
Long-acting muscarinic antagonist only 29 (6.7) 35 (7.7)
Long-acting beta2-agonist only 10 (2.3) 10 (2.2)
Inhaled glucocorticoids and long-acting muscarinic 5 (1.2) 9 (2.0)
antagonist
Inhaled glucocorticoids and long-acting beta2-agonist 71 (16.4) 89 (19.7)
Long-acting muscarinic antagonist and long-acting 25 (5.8) 23 (5.1)
beta2-agonist
Inhaled glucocorticoids, long-acting muscarinic 223 (51.5) 228 (50.4)
antagonist, and long-acting beta2-agonist
None 56 (12.9) 46 (10.2)
Entry criteria no. (%)
Acute COPD exacerbation requiring hospitalization or ED 216 (49.9) 238 (52.7)
visit within previous 12 mo
Systemic glucocorticoid or antibiotic use within previous 367 (84.8) 382 (84.5)
12 mo
Use of supplemental oxygen within previous 12 mo
All patients who met this criterion 198 (45.7) 222 (49.1)
Patients who met only this criterion 48 (11.1) 53 (11.7)

* Plusminus differences are means SD. There were no significant differences between the simvastatin and placebo
groups, and there was no imbalance regarding race between the two groups. COPD denotes chronic obstructive pul-
monary disease, ED emergency department, FEV1 forced expiratory volume in 1 second, and FVC forced vital capacity.
Spirometric measurements were available for 430 patients in the simvastatin group and 449 in the placebo group.
The Global Initiative for Chronic Obstructive Lung Disease (GOLD) staging system is used to assess the severity of
lung disease, with stage 4 indicating the most severe disease.

2206 n engl j med 370;23nejm.orgjune 5, 2014

The New England Journal of Medicine

Downloaded from nejm.org on December 26, 2016. For personal use only. No other uses without permission.
Copyright 2014 Massachusetts Medical Society. All rights reserved.
 Simvastatin for Prevention of Exacerbations in COPD

the simvastatin and placebo groups (0.631.56

events and 0.621.48 events, respectively; P>0.20 250
Placebo Statin
for all comparisons), except for nonfatal serious
adverse events involving the gastrointestinal tract, 200
which were more frequent with simvastatin than

No. of Participants
with placebo (in 30 patients [0.05 events per person- 150
year] vs. 17 patients [0.02 events per person-year],
P = 0.02) (Table 2). There were 28 deaths in the 100

simvastatin group and 30 in the placebo group

50
(P = 0.89 by life-table analysis).
0
PERMANENT DISCONTINUATION OF STUDY DRUG 0 0.01 1.50 1.51 3.00 3.01 4.50 4.51 6.00 6.01 7.50 7.51
AND STUDY WITHDRAWAL
No. of Exacerbations per Person-Yr
The most common reason for discontinuation of
the study drug was the use of incompatible med- Figure 2. Acute Exacerbations of Chronic Obstructive Pulmonary Disease
ications; other common reasons were a diagnosis per Person-Year, According to Study Group.
of diabetes and the FDA warning about additional The mean (SD) number of exacerbations per person-year were similar in
incompatible medications.28 A small number of the simvastatin and placebo groups: 1.361.61 exacerbations per person-
year and 1.391.73 exacerbations per person-year, respectively.
participants had myalgias or abnormal laboratory
values, but the rates did not differ according to
study-drug assignment (Table S6 in the Supplemen-
tary Appendix). A total of 36 participants (4.1%) 1.0
declined to continue participation for various Total No. Patients with Treatment
Proportion of Patients without Exacerbation

0.9 of Patients Failure (no./%)

reasons, all of which were unrelated to the study-
0.8 Placebo 447 292 (65.3)
drug assignment (Table S7 in the Supplementary Statin 430 293 (68.1)
Appendix). 0.7
0.6
P=0.34 by log-rank test
LIPID LEVELS 0.5
Total cholesterol, high-density lipoprotein and low- 0.4
density lipoprotein (LDL) cholesterol, and triglyc-
0.3
eride levels were similar in the simvastatin and Placebo
placebo groups at baseline; the LDL cholesterol 0.2
Statin
levels were 113.928.2 mg per deciliter (2.950.73 0.1
mmol per liter) and 114.427.6 mg per deciliter 0.0
0 100 200 300 400 500 600 700 800 900 1000 1100 1200
(2.960.71 mmol per liter), respectively. At 1 year
after randomization, high-density lipoprotein Follow-up (days)

cholesterol levels in the simvastatin group had

increased by 2.5 mg per deciliter (0.06 mmol per Figure 3. Effect of Simvastatin on the Time to the First Acute Exacerbation
of Chronic Obstructive Pulmonary Disease.
liter), whereas the levels in the placebo group had
There were no significant between-group differences in the time to the first
decreased by 0.5 mg per deciliter (0.01 mmol per
exacerbation. The median time to the first exacerbation was 223 days (95%
liter) (P = 0.01); LDL cholesterol levels had de- CI, 195 to 275) in the simvastatin group and 231 days (95% CI, 193 to 303)
creased by 33.2 mg per deciliter (0.86 mmol per in the placebo group.
liter) in the simvastatin group, as compared with
6.6 mg per deciliter (0.17 mmol per liter) in the
placebo group (P<0.001). (Complete data on lipid time to the first exacerbation, or the severity of ex-
levels are provided in Table S9 in the Supplemen- acerbations in patients with moderate-to-severe
tary Appendix.) COPD who were at high risk for exacerbations. Sim-
vastatin also had no effect on lung function or on
DISCUSSION general or disease-specific quality of life. The rates
of serious adverse events were low and similar in
In this large, prospective, multicenter, randomized the two study groups. These data show that daily
trial, we found that 40 mg of daily simvastatin had use of 40 mg of simvastatin has no role in prevent-
no effect on the frequency of exacerbations, the ing exacerbations in moderate-to-severe COPD.

n engl j med 370;23 nejm.org june 5, 2014 2207

The New England Journal of Medicine
Downloaded from nejm.org on December 26, 2016. For personal use only. No other uses without permission.
Copyright 2014 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

COPD and cardiovascular morbidity and mortal-

Table 2. Nonfatal Serious Adverse Events and Fatal Events, According to
Study Group. ity.24 A matched cohort study involving 76,232 pa-
tients showed that among those who received a
Simvastatin Placebo moderate statin dose (40 mg per day), the odds
Event (N=430) (N=447) P Value
ratios for influenza-related death and pneumonia-
Nonfatal serious adverse event
no. of events/
related death were significantly reduced.22 Similar
person-yr reductions in mortality were reported with statin
Acute exacerbation 0.32 0.32 0.99 use in population studies involving Japanese and
Norwegian patients with COPD.35,36 Keddissi et
Respiratory event
al. compared the annual decline in the FVC and
Pneumonia 0.01 <0.01 0.56
FEV1 according to statin use or nonuse in current
Chronic bronchitis 0.02 0.02 0.53 and former smokers who were treated at a Veter-
Other 0.01 0.01 0.75 ans Affairs (VA) medical center.23 Study partici-
Cancer pants who used statins had smaller declines, as
Lung cancer 0.02 0.03 0.81 compared with nonusers, in FEV1 (P<0.001) and
Other 0.03 0.03 0.21
FVC (P<0.001). The VA Normative Aging Study
showed that patients using statins had signifi-
Cardiovascular event
cantly less decline in FEV1, as compared with
Cardiac event 0.04 0.01 0.35 patients who did not use statins (P<0.001).20 The
Other 0.03 0.08 0.23 underuse of statins in persons with cardiac risk
Gastrointestinal tract event 0.05 0.02 0.02 factors who have been included in retrospective
Other 0.11 0.11 0.99 studies may account in part for the differences
Total 0.63 0.62 0.96 between our findings and those previously re-
ported.37
Fatal event no. of patients (%)
We chose to use simvastatin because it is one
Acute exacerbation 6 (1.4) 5 (1.1) 0.72
of the most effective HMG-CoA reductase inhibi-
Other respiratory event tors for lowering cardiovascular risk, it is widely
Pneumonia 1 (0.2) 1 (0.2) 0.99 available and affordable as a generic drug, and
Chronic bronchitis 0 2 (0.4) 0.16 the literature documenting its pleiotropic actions
Other 6 (1.4) 4 (0.9) 0.49 is extensive.38,39 The antilipid effects of simva
Cancer
statin are dose-dependent, as are the risks of
side effects such as myopathy.28 The antiinflam-
Lung cancer 3 (0.7) 4 (0.9) 0.73
matory effects of simvastatin are less dependent
Other 0 1 (0.2) 0.33
on the dose, with nearly maximal reductions in
Cardiovascular event serum C-reactive protein (CRP) levels at a dose
Heart 0 1 (0.2) 0.33 of 20 to 40 mg per day.40 We believed that anti-
Other 4 (0.9) 3 (0.7) 0.67 inflammatory effects were more important than
Gastrointestinal tract event 0 3 (0.7) 0.09
antilipid effects in producing the clinical bene-
fits we were seeking. Thus, the balance of risk
Other 8 (1.9) 6 (1.3) 0.56
to potential benefit, as well as reports of clear
Total 28 (6.5) 30 (6.7) 0.89
benefits with respect to mortality with simva
statin at the 40-mg daily dose, led us to favor
this lower dose over a dose of 80 mg per day
Our results differ from previously reported find- (which has greater antilipid effects).41,42
ings that statins prevent exacerbations, attenuate Our study has some limitations. First, it was
lung-function decline, and reduce mortality among terminated early for futility with respect to a
patients with COPD.20-24 A retrospective analysis treatment effect. However, a thorough examina-
of a large Canadian database showed that statin tion of the data by the STATCOPE data and safety
therapy in patients with COPD who did not have monitoring board and the investigators showed
cardiovascular disease was associated with large no signal for a delayed effect of simvastatin on
reductions in the rate of hospitalization due to the exacerbation rate. Furthermore, we did not

2208 n engl j med 370;23nejm.orgjune 5, 2014

The New England Journal of Medicine

Downloaded from nejm.org on December 26, 2016. For personal use only. No other uses without permission.
Copyright 2014 Massachusetts Medical Society. All rights reserved.
 Simvastatin for Prevention of Exacerbations in COPD

use a marker of systemic inflammation (serum evaluate the effect of simvastatin on the occurrence
CRP level) to screen patients for enrollment in of exacerbations in our study population.
the study, and this may have limited our ability In conclusion, simvastatin at a dose of 40 mg
to detect an effect of simvastatin in reducing the daily, in addition to usual care, did not reduce
rate of exacerbations.43 Also, the participants in the exacerbation rate or prolong the time to the
our study had moderate-to-severe airway ob- first exacerbation among patients with moderate-
struction, and it is unclear whether our observa- to-severe COPD who were at high risk for exacer-
tion that simvastatin was not beneficial would bations. Furthermore, simvastatin had no effect
apply to patients with less impairment. Finally, on lung function, quality of life, the rate of severe
new FDA recommendations regarding the inter- adverse events, or mortality. Thus, our data did not
actions of simvastatin with other drugs were an- show a therapeutic benefit of statins in patients
nounced during the study period. These recom- with moderate-to-severe COPD.
mendations hindered recruitment to a study that Supported by grants from the National Heart, Lung, and Blood
Institute of the National Institutes of Health (U10 HL074407,
was already impeded by high background statin U10 HL074408, U10HL074409, U10 HL074416, U10 HL074418,
use in clinical practice. Despite these recruitment U10 HL074422, U10 HL074424, U10 HL074428, U10 HL074431,
challenges, relatively few participants withdrew U10 HL074439, and U10 HL074441) and the Canadian Institutes
of Health Research (115074).
from the study or discontinued the study drug. The Disclosure forms provided by the authors are available with
low rate of withdrawal allowed us to conclusively the full text of this article at NEJM.org.

APPENDIX
The authors full names and degrees are as follows: Gerard J. Criner, M.D., John E. Connett, Ph.D., Shawn D. Aaron, M.D., Richard K.
Albert, M.D., William C. Bailey, M.D., Richard Casaburi, M.D., Ph.D., J. Allen D. Cooper, Jr., M.D., Jeffrey L. Curtis, M.D., Mark T.
Dransfield, M.D., MeiLan K. Han, M.D., Barry Make, M.D., Nathaniel Marchetti, D.O., Fernando J. Martinez, M.D., Dennis E.
Niewoehner, M.D., Paul D. Scanlon, M.D., Frank C. Sciurba, M.D., Steven M. Scharf, M.D., Ph.D., Don D. Sin, M.D., M.P.H., Helen
Voelker, B.A., George R. Washko, M.D., Prescott G. Woodruff, M.D., and Stephen C. Lazarus, M.D.
The authors affiliations are as follows: the Department of Pulmonary and Critical Care Medicine, Temple University School of
Medicine, Philadelphia (G.J.C., N.M.); Department of Biostatistics, School of Public Health (J.E.C.), Data Coordinating Center (J.E.C.,
H.V.), and Department of Pulmonary, Allergy, Critical Care, and Sleep Medicine (D.E.N.), University of Minnesota, Minneapolis, and
Department of Pulmonary and Critical Care Medicine, Mayo Clinic, Rochester (P.D.S.) all in Minnesota; Ottawa Hospital Research
Institute, University of Ottawa, Ottawa (S.D.A.), and Department of Medicine and University of British Columbia James Hogg Research
Centre, Institute for Heart and Lung Health, University of British Columbia, Vancouver (D.D.S.) both in Canada; Department of
Medicine, Denver Health Medical Center (R.K.A.), and Division of Pulmonary Critical Care and Sleep Medicine, National Jewish Health
(B.M.) both in Denver; Department of Pulmonary Allergy and Critical Care Medicine (W.C.B., M.T.D.) and the Pulmonary Section
(J.A.D.C.), University of Alabama at Birmingham and Birmingham Veterans Affairs Medical Center, Birmingham; HarborUCLA Re-
search and Education Institute, Los Angeles (R.C.); Department of Pulmonary Medicine, Veterans Affairs Medical Center and Univer-
sity of Michigan (J.L.C.), and Department of Pulmonary and Critical Care Medicine, University of Michigan Health System (M.K.H.,
F.J.M.) all in Ann Arbor; Department of Pulmonary, Allergy, and Critical Care Medicine, University of Pittsburgh Medical Center,
Pittsburgh (F.C.S.); Department of Pulmonary and Critical Care Medicine, University of Maryland, Baltimore (S.M.S.); Department of
Pulmonary and Critical Care Medicine, Brigham and Womens Hospital, Boston (G.R.W.); and Department of Pulmonary and Critical
Care Medicine, University of California, San Francisco, San Francisco (P.G.W., S.C.L.).

REFERENCES
1. Minio AM, Murphy SL, Xu J, Kochanek 5. Dalal AA, Christensen L, Liu F, Riedel the treatment of chronic obstructive pul-
KD. Deaths: final data for 2008. Natl Vital AA. Direct costs of chronic obstructive monary disease: a randomised controlled
Stat Rep 2011;59:1-126. pulmonary disease among managed care trial. Lancet 2003;361:449-56. [Erratum,
2. Hurst JR, Vestbo J, Anzueto A, et al. patients. Int J Chron Obstruct Pulmon Dis Lancet 2003;361:1660.]
Susceptibility to exacerbation in chronic 2010;5:341-9. 10. Jialal I, Stein D, Balis D, Grundy SM,
obstructive pulmonary disease. N Engl J 6. Druss BG, Marcus SC, Olfson M, Pincus Adams-Huet B, Devaraj S. Effect of hydroxy
Med 2010;363:1128-38. HA. The most expensive medical conditions methyl glutaryl coenzyme a reductase in-
3. Seemungal TA, Donaldson GC, Paul in America. Health Aff (Millwood) 2002; hibitor therapy on high sensitive C-reactive
EA, Bestall JC, Jeffries DJ, Wedzicha JA. 21(4):105-11. protein levels. Circulation 2001;103:1933-5.
Effect of exacerbation on quality of life in 7. Tashkin DP, Celli B, Senn S, et al. 11. Johnson BA, Iacono AT, Zeevi A, Mc-
patients with chronic obstructive pulmo- A4-year trial of tiotropium in chronic ob- Curry KR, Duncan SR. Statin use is asso-
nary disease. Am J Respir Crit Care Med structive pulmonary disease. N Engl J Med ciated with improved function and survival
1998;157:1418-22. 2008;359:1543-54. of lung allografts. Am J Respir Crit Care
4. Soler-Catalua JJ, Martnez-Garca MA, 8. Albert RK, Connett J, Bailey WC, et al. Med 2003;167:1271-8.
Romn Snchez P, Salcedo E, Navarro M, Azithromycin for prevention of exacerba- 12. Mortensen EM, Copeland LA, Pugh MJ,
Ochando R. Severe acute exacerbations tions of COPD. N Engl J Med 2011;365:689- et al. Impact of statins and ACE inhibitors
and mortality in patients with chronic ob- 98. [Erratum, N Engl J Med 2012;366:1356.] on mortality after COPD exacerbations.
structive pulmonary disease. Thorax 2005; 9. Calverley P, Pauwels R, Vestbo J, et al. Respir Res 2009;10:45.
60:925-31. Combined salmeterol and fluticasone in 13. Kruger P, Bailey M, Bellomo R, et al.

n engl j med 370;23nejm.orgjune 5, 2014 2209

The New England Journal of Medicine
Downloaded from nejm.org on December 26, 2016. For personal use only. No other uses without permission.
Copyright 2014 Massachusetts Medical Society. All rights reserved.
 Simvastatin for Prevention of Exacerbations in COPD

Amulticenter randomized trial of atorva 24. Mancini GB, Etminan M, Zhang B, 34. Lan KKG, DeMets DL. Design and
statin therapy in intensive care patients Levesque LE, FitzGerald JM, Brophy JM. analysis of group sequential tests based on
with severe sepsis. Am J Respir Crit Care Reduction of morbidity and mortality by the type 1 error spending rate function.
Med 2013;187:743-50. statins, angiotensin-converting enzyme in- Biometrika 1987;74:149-54.
14. Kinlay S, Schwartz GG, Olsson AG, et hibitors, and angiotensin receptor blockers 35. Ishida W, Kajiwara T, Ishii M, et al.
al. High-dose atorvastatin enhances the in patients with chronic obstructive pul- Decrease in mortality rate of chronic ob-
decline in inflammatory markers in pa- monary disease. J Am Coll Cardiol 2006; structive pulmonary disease (COPD) with
tients with acute coronary syndromes in 47:2554-60. statin use: a population-based analysis in
the MIRACL study. Circulation 2003;108: 25. Lee TM, Chen CC, Shen HN, Chang NC. Japan. Tohoku J Exp Med 2007;212:265-
1560-6. Effects of pravastatin on functional capac- 73.
15. Inoue I, Goto S, Mizotani K, et al. ity in patients with chronic obstructive 36. Syseth V, Brekke PH, Smith P, Om-
Lipophilic HMG-CoA reductase inhibitor pulmonary disease and pulmonary hyper- land T. Statin use is associated with re-
has an anti-inflammatory effect: reduction tension. Clin Sci (Lond) 2009;116:497-505. duced mortality in COPD. Eur Respir J
of MRNA levels for interleukin-1beta, inter 26. Rabe KF, Hurd S, Anzueto A, et al. 2007;29:279-83.
leukin-6, cyclooxygenase-2, and p22phox Global strategy for the diagnosis, man- 37. Andell P, Koul S, Martinsson A, et al.
by regulation of peroxisome proliferator- agement, and prevention of chronic ob- Impact of chronic obstructive pulmonary
activated receptor alpha (PPARalpha) in structive lung disease: GOLD executive disease on morbidity and mortality after
primary endothelial cells. Life Sci 2000; summary. Am J Respir Crit Care Med myocardial infarction. Open Heart 2014;1:
67:863-76. 2007;176:532-55. e000002 (http://openheart.bmj.com/content/
16. Hothersall E, McSharry C, Thomson 27. Detection, evaluation, and treatment 1/1/e000002.full.pdf).
NC. Potential therapeutic role for statins of high blood cholesterol in adults (Adult 38. Heart Protection Study Collaborative
in respiratory disease. Thorax 2006;61: Treatment Panel III): final report 2014. Group. MRC/BHF Heart Protection Study
729-34. Bethesda, MD: National Heart, Lung, and of cholesterol lowering with simvastatin
17. Ferns GA. Differential effects of statins Blood Institute (http://www.nhlbi.nih.gov/ in 20,536 high-risk individuals: a random
on serum CRP levels: implications of re- guidelines/cholesterol/atp3full.pdf). ised placebo-controlled trial. Lancet 2002;
cent clinical trials. Atherosclerosis 2003; 28. FDA drug safety communication: on- 360:7-22.
169:349-51. going safety review of high-dose Zocor 39. Collins R, Armitage J, Parish S, Sleigh P,
18. Bellosta S, Via D, Canavesi M, et al. (simvastatin) and increased risk of muscle Peto R. MRC/BHF Heart Protection Study
HMG-CoA reductase inhibitors reduce injury. March 2014 (http://www.fda.gov/ of cholesterol-lowering with simvastatin in
MMP-9 secretion by macrophages. Arterio Drugs/DrugSafety/PostmarketDrugSafety 5963 people with diabetes: a randomised
scler Thromb Vasc Biol 1998;18:1671-8. InformationforPatientsandProviders/ placebo-controlled trial. Lancet 2003;361:
19. Balk EM, Lau J, Goudas LC, et al. Ef- ucm204882.htm#TableofSimvastatinDose 2005-16.
fects of statins on nonlipid serum mark- Limitations). 40. Li JJ, Chen MZ, Chen X, Fang CH.
ers associated with cardiovascular disease: 29. Barr JT, Schumacher GE, Freeman S, Rapid effects of simvastatin on lipid pro-
a systematic review. Ann Intern Med LeMoine M, Bakst AW, Jones PW. Ameri- file and C-reactive protein in patients
2003;139:670-82. can translation, modification, and valida- with hypercholesterolemia. Clin Cardiol
20. Alexeeff SE, Litonjua AA, Sparrow D, tion of the St. Georges Respiratory Ques- 2003;26:472-6.
Vokonas PS, Schwartz J. Statin use reduces tionnaire. Clin Ther 2000;22:1121-45. 41. Scandinavian Simvastatin Survival
decline in lung function: VA Normative 30. Ware JE Jr, Sherbourne CD. The MOS Study Group. Randomised trial of cho-
Aging Study. Am J Respir Crit Care Med 36-item short-form health survey (SF-36). lesterol lowering in 4444 patients with
2007;176:742-7. I. Conceptual framework and item selection. coronary heart disease: the Scandinavian
21. Blamoun AI, Batty GN, DeBari VA, Med Care 1992;30:473-83. Simvastatin Survival Study (4S). Lancet
Rashid AO, Sheikh M, Khan MA. Statins 31. Wyrwich KW, Finn SD, Tierney WM, 1994;344:1383-9.
may reduce episodes of exacerbation and Kroenke K, Babu AN, Wolinsky FD. Clini- 42. MAAS Investigators. Effect of simva
the requirement for intubation in patients cally important changes in health-related statin on coronary atheroma: the Multi-
with COPD: evidence from a retrospective quality of life for patients with chronic centre Anti-Atheroma Study (MAAS). Lan-
cohort study. Int J Clin Pract 2008;62:1373-8. obstructive pulmonary disease. J Gen In- cet 1994;344:633-8.
22. Frost FJ, Petersen H, Tollestrup K, tern Med 2003;18:196-202. 43. Lahousse L, Loth DW, Joos GF, et al.
Skipper B. Influenza and COPD mortality 32. Miller MR, Hankinson J, Brusasco V, Statins, systemic inflammation and risk
protection as pleiotropic, dose-dependent et al. Standardisation of spirometry. Eur of death in COPD: the Rotterdam study.
effects of statins. Chest 2007;131:1006-12. Respir J 2005;26:319-38. Pulm Pharmacol Ther 2013;26:212-7.
23. Keddissi JI, Younis WG, Chbeir EA, 33. Hankinson JL, Odencrantz JR, Fedan Copyright 2014 Massachusetts Medical Society.
Daher NN, Dernaika TA, Kinasewitz GT. KB. Spirometric reference values from a
The use of statins and lung function in sample of the general U.S. population.
current and former smokers. Chest 2007; Am J Respir Crit Care Med 1999;159:179-
132:1764-71. 87.

receive immediate notification when an article

is published online first

To be notified by e-mail when Journal articles

are published Online First, sign up at NEJM.org.

2210 n engl j med 370;23nejm.orgjune 5, 2014

The New England Journal of Medicine

Downloaded from nejm.org on December 26, 2016. For personal use only. No other uses without permission.
Copyright 2014 Massachusetts Medical Society. All rights reserved.