Glucocorticoid use and risk of atrial fibrillation or flutter: a population-based, case-control study.

Abstract

Glucocorticoid use is associated with increased risk of myocardial infarction, stroke, and heart failure, but data are limited on the risk of atrial fibrillation or flutter. We examined whether glucocorticoid use is associated with the risk of atrial fibrillation or flutter.
For this population-based, case-control study, we identified all patients with a first hospital diagnosis of atrial fibrillation or flutter from January 1, 1999, through December 31, 2005, in Northern Denmark (population, 1.7 million). For each case we selected 10 population controls matched by age and sex. We obtained data on glucocorticoid prescriptions within 60 days (current users) or longer before the index date (former users), comorbidity, and medications from medical databases. We used conditional logistic regression to compute odds ratios (ORs), controlling for potential confounders.
Among 20,221 patients with atrial fibrillation or flutter, 1288 (6.4%) were current glucocorticoid users and 2375 (11.7%) were former users. Among 202,130 population controls, 5245 (2.6%) were current glucocorticoid users and 19 940 (9.9%) were former users. Current glucocorticoid use was associated with an increased risk of atrial fibrillation or flutter compared with never use (adjusted OR, 1.92; 95% confidence interval [CI], 1.79-2.06). Among new glucocorticoid users, the adjusted OR was 3.62 (95% CI, 3.11-4.22) and among long-term users it was 1.66 (95% CI, 1.53-1.80). The increased risk remained robust in patients with and without pulmonary and cardiovascular diseases. Former glucocorticoid use was not associated with increased risk (adjusted OR, 1.00; 95% CI, 0.96-1.06).
Current glucocorticoid use was associated with an almost 2-fold increased risk of atrial fibrillation or flutter.

Full text

Page 1

ORIGINAL INVESTIGATION
Glucocorticoid Use and Risk of Atrial Fibrillation
or Flutter
A Population-Based, Case-Control Study
Christian Fynbo Christiansen, MD; Steffen Christensen, MD; Frank Mehnert, MSc; Steven R. Cummings, MD;
Roland D. Chapurlat, MD, PhD; Henrik Toft Sørensen, MD, PhD, DMSc
Background: Glucocorticoid use is associated with
increased risk of myocardial infarction, stroke, and
heart failure, but data are limited on the risk of atrial
fibrillation or flutter. We examined whether glucocorti-
coid use is associated with the risk of atrial fibrillation
or flutter.
Methods: For this population-based, case-control
study, we identified all patients with a first hospital
diagnosis of atrial fibrillation or flutter from January 1,
1999, through December 31, 2005, in Northern Den-
mark (population, 1.7 million). For each case we
selected 10 population controls matched by age and
sex. We obtained data on glucocorticoid prescriptions
within 60 days (current users) or longer before the
index date (former users), comorbidity, and medica-
tions from medical databases. We used conditional
logistic regression to compute odds ratios (ORs), con-
trolling for potential confounders.
Results:Among20221patientswithatrialfibrillationor
flutter,1288(6.4%)werecurrentglucocorticoidusersand
2375 (11.7%) were former users. Among 202130 popu-
lation controls, 5245 (2.6%) were current glucocorticoid
users and 19940 (9.9%) were former users. Current glu-
cocorticoidusewasassociatedwithanincreasedriskofatrial
fibrillation or flutter compared with never use (adjusted
OR,1.92;95%confidenceinterval[CI],1.79-2.06).Among
newglucocorticoidusers,theadjustedORwas3.62(95%
CI,3.11-4.22)andamonglong-termusersitwas1.66(95%
CI, 1.53-1.80). The increased risk remained robust in pa-
tientswithandwithoutpulmonaryandcardiovasculardis-
eases. Former glucocorticoid use was not associated with
increased risk (adjusted OR,1.00; 95% CI, 0.96-1.06).
Conclusion: Current glucocorticoid use was associ-
ated with an almost 2-fold increased risk of atrial fibril-
lation or flutter.
Arch Intern Med. 2009;169(18):1677-1683
G
rheumatoid arthritis (RA), connective tis-
sue diseases, inflammatory bowel dis-
eases,chronichepatitis,glomerulonephri-
tis,andmalignantneoplasms,aswellasin
organtransplantations.1Glucocorticoiduse
has several adverse effects that are estab-
lishedriskfactorsforatherosclerosis,such
as hypertension, sodium and fluid reten-
tion, diabetes mellitus, and dyslipid-
emia.2,3Treatment with high-dose gluco-
corticoidshasbeenreportedtoincreasethe
risk of myocardial infarction, stroke, and
heart failure.4Atrial fibrillation is a com-
mon cardiac arrhythmia whose preva-
lenceisabove10%inmen80yearsorolder.5
Itisassociatedwithincreasedmortalityand
morbidity,mainlybecauseofa3-to4-fold
increasedriskofischemicstroke.6Theex-
isting literature with regard to glucocorti-
coiduseandatrialfibrillationconsistsof2
LUCOCORTICOIDS ARE
widely used in the treat-
mentofinflammatoryand
autoimmune diseases,
such as chronic obstruc-
tive pulmonary disease (COPD), asthma,
casereportsand2case-controlstudies.Both
casereportsdescribedepisodesofatrialfi-
brillation after methylprednisolone pulse
therapy.7,8The2case-controlstudies,which
involved385and468atrialfibrillationcases
fromtheNetherlandsandtheUnitedKing-
dom,respectively,foundanincreasedrisk
of atrial fibrillation among users of sys-
temicglucocorticoids.9,10However,thein-
terpretation of these studies was ham-
pered by restriction to COPD patients,10
incomplete participation,9and relatively
small sample size which yielded statisti-
cally imprecise risk estimates for
subgroups.9
Because atrial fibrillation is associated
withseverecomplicationsandglucocorti-
coids are widely used, an association be-
tweenglucocorticoidsandatrialfibrillation
has major public health implications. To
address the limitations in the existing lit-
erature, we therefore conducted a large,
population-based,case-controlstudytoex-
amine whether and to what extent use of
systemicglucocorticoidsisassociatedwith
increasedriskofatrialfibrillationorflutter.
Author Affiliations:
Departments of Clinical
Epidemiology
(Drs Christiansen, Christensen,
and Sørensen and Mr Mehnert)
and Anesthesiology and
Intensive Care Medicine
(Dr Christiansen), Aarhus
University Hospital, Aarhus,
Denmark; San Francisco
Coordinating Center, California
(Dr Cummings); and Institut
National de la Sante ´ et la
Recherche Me ´dicale Research
Unit 831, Universite ´ de Lyon,
and Department of Orthopedics
and Rheumatology, Ho ˆpital
Edouard Herriot, Lyon, France
(Dr Chapurlat).
(REPRINTED) ARCH INTERN MED/VOL 169 (NO. 18), OCT 12, 2009WWW.ARCHINTERNMED.COM
1677
©2009 American Medical Association. All rights reserved.
at Capes Consortia, on October 12, 2009 www.archinternmed.comDownloaded from

Page 2

METHODS
We conducted this population-based, case-control study in
Northern Denmark, within a population of 1.7 million (ap-
proximately 30% of the Danish population). The Danish
National Health Service provides free tax-funded medical
care for all Danish residents and partially reimburses costs of
most physician-prescribed drugs. Only a few cardiologists
work outside public hospitals in Denmark, and most
patients with atrial fibrillation or flutter have their condi-
tions diagnosed during hospital admission or at hospital out-
patient clinics. The unique civil registration number
assigned to every Danish citizen and included in all Danish
medical databases allowed us to electronically link several
medical databases.11For this study we used data from 3 dif-
ferent databases (the Danish National Registry of Patients
[NRP], the prescription database of the region, and the Dan-
ish Civil Registration System [CRS]), merged into a research
database at Aarhus University Hospital.12,13
CASES OF ATRIAL FIBRILLATION
OR FLUTTER
We used the NRP to identify all patients with an incident
hospital diagnosis of atrial fibrillation or flutter from January
1, 1999, through December 31, 2005. We did not include
patients with any earlier diagnosis of atrial fibrillation or
flutter registered in the NRP. The NRP contains data with
regard to all discharges from all nonpsychiatric hospitals
since 1977 and with regard to emergency department and
outpatient clinic visits since 1995. It includes information
with regard to civil registration numbers of patients, dates of
admission and discharge, and up to 20 discharge diagnoses,
coded by physicians in accordance with the International
Classification of Diseases, Eighth Revision (ICD-8),14until the
end of 1993 and the International Statistical Classification of
Diseases, 10th Revision (ICD-10), thereafter.15
The ICD-8 codes used to identify patients with atrial
fibrillation or flutter were 427.93 and 427.94, and the
ICD-10 code was I48.9. The single ICD-10 code includes
both atrial fibrillation and flutter; however, it has been
reported that more than 90% of patients registered with
atrial fibrillation and flutter have atrial fibrillation and 5%
had atrial flutter only.16We considered it appropriate to
include both atrial fibrillation and flutter because they are
often seen in the same patients and share risk factors and, to
some extent, pathophysiologic factors.17,18The positive pre-
dictive value of atrial fibrillation or flutter has been reported
to be as high as 97% in the NRP.19We also obtained data
from the NRP with regard to hospital procedure codes of
cardioversions (treatment code BFFA0 in the Danish treat-
ment classification) within the year after diagnosis of atrial
fibrillation or flutter.
POPULATION CONTROLS
For each atrial fibrillation or flutter case, we used the
CRS to select 10 population controls matched on age and
sex. The CRS is updated daily and contains data on civil reg-
istration number, residence, migration, vital status (dead or
alive), and exact date of death for all Danish residents since
1968.11The controls were selected by means of risk set sam-
pling and assigned an index date identical to the hospital
admission date for atrial fibrillation or flutter of the matched
case.
GLUCOCORTICOID USE
All pharmacies in the region are equipped with electronic ac-
counting systems, established to secure partial reimburse-
ment of prescription expenses from the National Health Ser-
vice.13The prescription database contains information with
regard to the civil registration numbers of customers, the type
and amount of drugs prescribed (according to the Anatomic
TherapeuticChemical[ATC]classificationsystem),andthepre-
scription fill dates. We used the prescription database to iden-
tify any prescription for systemic glucocorticoids reimbursed
before the date of admission for atrial fibrillation or flutter or
index date. We included prednisolone, prednisone, methyl-
prednisolone, dexamethasone, and hydrocortisone (the ATC
codesareprovidedintheeAppendix,http://www.archinternmed
.com),whichareusuallyprescribedfororaladministrationfor
outpatient use. We did not include triamcinolone acetonide,
triamcinolonehexacetonide,orbetamethasone,whichinDen-
mark are used only for local injections (eg, in the treatment of
bursitis and synovitis).
We defined current users of glucocorticoids as patients who
filledtheirmostrecentprescriptionforglucocorticoidswithin60
days before the index date.20Because some adverse effects may
arise shortly after initiation of therapy, and because inclusion of
prevalentusersmayunderestimatethesecomplications,21wefur-
ther categorized current users of glucocorticoids into 2 groups:
newusers,definedashavingfilledtheirfirst-everprescriptionfor
glucocorticoids within 60 days before the index date, and long-
termusers,definedascurrentuserswhofilledtheirfirstprescrip-
tion more than 60 days before the index date. Patients who had
filledtheirmostrecentprescriptionmorethan60daysbeforethe
index date were defined as former users. Never users were de-
fined as patients with no filled prescriptions for glucocorticoids
registered in the prescription database.
Amongusersofprednisoloneandprednisone,weexamined
adose-responserelationwiththeriskofatrialfibrillationorflut-
ter.Thesedrugshaveclearindicationsandaccountfor80%ofpre-
scribedsystemicglucocorticoids.Weobtaineddatawithregard
todosesoftabletsofprednisoloneandprednisoneprescribedmost
recentlyrelativetotheadmissionorindexdateandusedtheseto
computeoddsratios(ORs)forcurrentlow-doseuse(2.5-mgor
5-mg tablets) and for current high-dose use (25-mg tablets).
COMORBIDITY
A number of risk factors for atrial fibrillation or flutter also may
beassociatedwithuseofglucocorticoids.Wethereforeobtained
datafromtheNRPonanyprevioushospitaldiagnosissince1977
ofcardiovasculardiseases,COPDandasthma,diabetes,renalfail-
ure, inflammatory bowel disease, connective tissue disease and
RA, liver disease and chronic pancreatitis, acute alcohol intoxi-
cation, alcoholism-related diseases, hyperthyroidism, and can-
cer. Associated ICD codes are provided in the eAppendix.
We used the prescription database to obtain data on ever
useofantithyroiddrugs,thyroidhormones,respiratorydrugs,
and cardiovascular drugs (with the inclusion of angiotensin-
convertingenzymeinhibitors,?-blockingagents,low-doseas-
pirin, statins, calcium antagonists, and other antihyperten-
sives,diuretics,andnitrates)asmarkersofthyroid,pulmonary,
or cardiovascular diseases. Any use of nonsteroidal anti-
inflammatorydrugsorselectivecyclooxygenase2inhibitorswas
also included as a potential confounder. Their ATC codes are
provided in the eAppendix. To identify patients with atrial fi-
brillationorfluttertreatedonlybytheirgeneralphysician(and
thereforenotregisteredintheNRP),weobtaineddatawithre-
gard to previous use of digoxin and vitamin K antagonists by
all study participants before their index dates.
(REPRINTED) ARCH INTERN MED/VOL 169 (NO. 18), OCT 12, 2009 WWW.ARCHINTERNMED.COM
1678
©2009 American Medical Association. All rights reserved.
at Capes Consortia, on October 12, 2009 www.archinternmed.com Downloaded from

Page 3

STATISTICAL ANALYSIS
WeusedconditionallogisticregressiontoestimateORsforatrial
fibrillation or flutter among current and former users of glu-
cocorticoids compared with never users, controlling for sex,
age group, comorbidity, and drug use. Because we used risk
setsamplingofcontrols,theORsareunbiasedestimatesofthe
incidence rate ratio. To not include patients with atrial fibril-
lationorfluttertreatedbytheirgeneralphysicianwithoutbeing
previously hospitalized, we restricted the primary analyses to
those cases and controls who had never before the index date
filled a prescription for digoxin or a vitamin K antagonist. In a
secondaryanalysisweincludedprevioususersoftheseagents.
Another secondary analysis was limited to cases with atrial fi-
brillation or flutter recorded as the first listed diagnosis in the
discharge summary and their controls. (The first listed diag-
nosis is the main reason for hospital admission or outpatient
or emergency department visit.)
We repeated the analysis with the inclusion of only cases
treated with cardioversion within 1 year after the index date
and their controls, as a measure of severity of atrial fibrillation
or flutter. To reduce potential confounding by indication, we
also performed the analyses in prespecified subgroups of pa-
tients with and without cardiovascular diseases or cardiovas-
culardruguse,COPDandasthmaorrespiratorydruguse,and
connectivetissuediseaseandRA.Wekeptthematchinginthese
subgroups and performed conditional logistic regression by
means of only cases and controls with the subgroup disease.
Toassesstheimpactofsurveillancebiasthatstemmedfrom
an increased number of hospitalizations among users of glu-
cocorticoids,werepeatedtheanalysisinsubgroupsdefinedby
the number of hospital admissions in the year preceding the
index date (0, 1, or ?2 admissions). We also performed strati-
fied analysis in age groups.
StatisticalanalyseswereperformedwithSASstatisticalsoft-
ware(version9.1,SASInstituteInc,Cary,NorthCarolina).The
studywasapprovedbytheDanishDataProtectionAgency(rec-
ord 2004-41-4693) and the Aarhus University Registry Board.
SENSITIVITY ANALYSIS
We performed a sensitivity analysis to quantify the influence
of potential unmeasured confounding on the observed asso-
ciation by means of a rule-out approach.22We estimated how
strongly a single unmeasured binary confounder would need
to be associated with glucocorticoid use and atrial fibrillation
or flutter to fully explain our findings and illustrated this as-
sociation graphically. We assumed that the prevalence of such
a confounder was as common as smoking (30.0% of the popu-
lation) and that 2.5% of the population used glucocorticoids.
Ifatrialfibrillationorflutterbeganbeforethedateofthefirst
diagnosis,wewouldalsoexpectanincreasedriskinpatientswho
filled prescriptions of glucocorticoids more than 60 days before
the index date compared with never users. To examine the sen-
sitivity of the estimates to differences in exposure definition, we
therefore repeated the analysis for exposure (filled glucocorti-
coid prescription) within 30, 60, 90, 120, and 180 days.
RESULTS
DESCRIPTIVE DATA
We identified 20221 patients with an incident hospital
diagnosisofatrialfibrillationorflutterand202130popu-
lation controls, with the exclusion from both groups of
persons with previous prescriptions for digoxin or vita-
min K antagonists (Table 1). Among the cases, 17701
(8.76%) had atrial fibrillation or flutter diagnosed dur-
ing hospital admission and 2520 (1.25%) at an outpa-
tient clinic or emergency department.
Among the 20221 cases, 1288 (6.4%) were current
usersofglucocorticoids(0useddexamethasone,41[0.2%]
used methylprednisolone, 845 [4.2%] used predniso-
lone, 397 [2.0%] used prednisone, and 5 [0.02%] used
hydrocortisone). A total of 2375 cases (11.7%) were
former glucocorticoid users. Among the 202130 con-
trols, 5245 (2.6%) were current glucocorticoid users,
19940 (9.9%) were former users, and 176945 (87.5%)
were never users (Table 1).
Among the cases, 35.1% had been previously diag-
nosedashavingcardiovasculardiseasesotherthanatrial
fibrillation or flutter, compared with 18.5% of the con-
trols.Useofcardiovasculardrugswas,asexpected,more
frequentamongthecasesthanamongthecontrols(69.6%
vs 49.9%). In addition, COPD and asthma, diabetes, hy-
perthyroidism,andconnectivetissuediseaseandRAwere
more common among the cases than among the popu-
lation controls (Table 1).
OVERALL RISK OF ATRIAL FIBRILLATION
OR FLUTTER
Currentuseofglucocorticoidswasassociatedwithanin-
creasedriskofhospitalizationforatrialfibrillationorflut-
ter, with an adjusted OR of 1.92 (95% CI, 1.79-2.06)
(Table 2). Former glucocorticoid use was not associ-
ated with the risk of atrial fibrillation or flutter (ad-
justed OR,1.00; 95% CI, 0.96-1.06). The approximately
2-fold risk increase was seen in all age groups except for
the patients younger than 50 years, for whom the esti-
mate was somewhat lower but imprecise (0-49 years:
OR,1.23;95%CI,0.63-2.43;50-59years:OR,1.77;95%
CI, 1.24-2.50; 60-69 years: OR,2.17; 95% CI, 1.82-2.60;
70-79 years: OR,1.95; 95% CI, 1.74-2.19; ?80 years:
OR,1.82; 95% CI, 1.63-2.04).
Inclusion of the 8100 cases with a history of digoxin
orvitaminKantagonistuseandtheircontrolsintheanaly-
sis had no major impact on the risk estimate for atrial
fibrillation or flutter among current glucocorticoid us-
ers (OR,1.74; 95% CI, 1.64-1.85). The adjusted OR for
atrialfibrillationorflutteramongthe9901caseswithatrial
fibrillation or flutter listed as the first diagnosis in the
discharge record was 1.60 (95% CI, 1.43-1.78).
FURTHER ANALYSIS
Amongthenewusersofglucocorticoids,theadjustedOR
was 3.62 (95% CI, 3.11-4.22), and among long-term us-
ers,theadjustedORwas1.66(95%CI,1.53-1.80).Among
current users of high-dose prednisolone or prednisone
tablets, the adjusted OR was 4.03 (95% CI, 3.44-4.72),
and the adjusted OR in current users of low-dose tablets
was 1.78 (95% CI, 1.65-1.92), compared with never us-
ers of these drugs.
In the 1793 cases (8.9%) treated with cardioversion
within the year after their first atrial fibrillation or flut-
ter diagnosis, the adjusted OR for current glucocorti-
coid use was 1.86 (95% CI, 1.41-2.45). The risk of atrial
(REPRINTED) ARCH INTERN MED/VOL 169 (NO. 18), OCT 12, 2009WWW.ARCHINTERNMED.COM
1679
©2009 American Medical Association. All rights reserved.
at Capes Consortia, on October 12, 2009 www.archinternmed.com Downloaded from

Page 4

fibrillation or flutter among current users of glucocorti-
coids remained elevated in analyses restricted to cases
and controls with and without COPD and asthma and
cardiovascular diseases (Table 3). The risk was also in-
creasedamongpatientswithconnectivetissuediseaseand
RA, although the associated estimates were statistically
imprecise. Stratification by the number of hospital ad-
missionsintheyearbeforetheindexdatehadonlyami-
norimpactontheadjustedrelativeriskestimates(noad-
missions: OR,1.81; 95% CI, 1.65-2.00; 1 admission:
OR,1.33; 95% CI, 0.93-1.89; and ?2 admissions: OR,
1.71; 95% CI, 1.29-2.27).
Table 1. Characteristics of Cases With Atrial Fibrillation or Flutter and Controls From Northern Denmark, 1999-2005
Characteristics
Cases, No. (%)
(n=20221)
Population Controls, No. (%)
(n=202130)
Sex
Male
Female
Age group, y
0-49
50-59
60-69
70-79
?80
Use of systemic glucocorticoids
Current use
Former use
Never use
Cardiovascular diseases
Previous hospital diagnosis, with the inclusion of hypertension
Any use of cardiovascular drugs
Angiotensin-converting enzyme inhibitors
?-Blockers
Low-dose aspirin
Statins
Calcium antagonists
Diuretics
Nitrates
Other antihypertensives
Renal failure
Diabetes mellitus, previous hospital diagnosis
COPD and asthma
Previous hospital diagnosis
Any use of respiratory drugs
Connective tissue disease and RA
Any use of nonsteroidal anti-inflammatory drugs or cyclooxygenase 2 inhibitors
Inflammatory bowel disease
Cancer
Liver disease and chronic pancreatitis
Concurrent acute alcohol intoxication
Alcohol-related diseases
Hyperthyroidism
Any use of antithyroid drugs
Any use of thyroid hormones
10948 (54.1)
9273 (45.9)
109450 (54.1)
92680 (45.9)
1157 (5.7)
2338 (11.6)
3969 (19.6)
6230 (30.8)
6527 (32.3)
11618 (5.7)
23291 (11.5)
39894 (19.7)
61927 (30.6)
65400 (32.4)
1288 (6.4)
2375 (11.7)
16558 (81.9)
14581 (72.1)
7100 (35.1)
14072 (69.6)
5131 (25.4)
6174 (30.5)
3911 (19.3)
2010 (9.9)
1566 (7.7)
10058 (49.7)
3817 (18.9)
542 (2.7)
392 (1.9)
1238 (6.1)
4755 (23.5)
2219 (11.0)
4338 (21.5)
818 (4.0)
9954 (49.2)
179 (0.9)
2582 (12.8)
1880 (9.3)
162 (0.8)
412 (2.0)
602 (3.0)
539 (2.7)
666 (3.3)
5245 (2.6)
19940 (9.9)
176945 (87.5)
103971 (51.4)
37314 (18.5)
100915 (49.9)
34440 (17.0)
33385 (16.5)
27021 (13.4)
13511 (6.7)
7401 (3.7)
69671 (34.5)
21489 (10.6)
3517 (1.7)
1218 (0.6)
7566 (3.7)
30606 (15.1)
12231 (6.1)
27849 (13.8)
5981 (3.0)
89125 (44.1)
1344 (0.7)
19457 (9.6)
10175 (5.0)
924 (0.5)
2461 (1.2)
3956 (2.0)
3355 (1.7)
6629 (3.3)
Abbreviations: COPD, chronic obstructive pulmonary disease; RA, rheumatoid arthritis.
Table 2. Association Between Glucocorticoid Use and Atrial Fibrillation or Flutter
Glucocorticoid Use
Never use
Current use
New users
Long-term users
Former use
No. of Cases/Controls
16558/176945
1288/5245
280/628
1008/4617
2375/19940
Odds Ratio (95% Confidence Interval)
Unadjusted
1 [Reference]
2.66 (2.49-2.83)
4.72 (4.09-5.46)
2.37 (2.21-2.55)
1.28 (1.23-1.34)
Adjusteda
1 [Reference]
1.92 (1.79-2.06)
3.62 (3.11-4.22)
1.66 (1.53-1.80)
1.00 (0.96-1.06)
aAdjusted for a previous diagnosis of cardiovascular diseases, chronic obstructive pulmonary disease and asthma, diabetes mellitus, renal failure, inflammatory
bowel disease, connective tissue disease and rheumatoid arthritis, cancer, liver disease and chronic pancreatitis, acute alcohol intoxication, alcoholism-related
diseases, hyperthyroidism, and use of nonsteroidal anti-inflammatory drugs or cyclooxygenase 2 inhibitors and cardiovascular, respiratory, and thyroid drugs.
(REPRINTED) ARCH INTERN MED/VOL 169 (NO. 18), OCT 12, 2009 WWW.ARCHINTERNMED.COM
1680
©2009 American Medical Association. All rights reserved.
at Capes Consortia, on October 12, 2009 www.archinternmed.comDownloaded from

Page 5

SENSITIVITY ANALYSIS
Thesensitivityanalysisrevealedthatanunmeasuredcon-
founder that is 4 times more frequent among users than
nonusers of glucocorticoids would need to increase the
riskofatrialfibrillationorflutterbyafactorof9ormore
toexplainfullyourfindingofanadjustedORof1.92(95%
CI, 1.79-2.06), if no increased risk actually existed
(Figure). The risk estimates decreased only slightly as
a result of increasing of the exposure time window (30
days:adjustedOR,2.01;95%CI,1.85-2.19;60days:ad-
justedOR,1.92;95%CI,1.79-2.06;90days:adjustedOR,
1.76; 95% CI, 1.65-1.88; 120 days: adjusted OR, 1.69;
95%CI,1.58-1.80;120days,adjustedOR,1.59;95%CI,
1.50-1.69).
COMMENT
In this large, population-based, case-control study, we
found that use of systemic glucocorticoids was associ-
ated with an almost doubled risk of atrial fibrillation or
flutter. The risk was 4 times higher among new users of
glucocorticoidscomparedwithneverusers,andthefind-
ing was robust in patients with and without COPD or
asthma and cardiovascular diseases.
Ourfindingsextendthedatafromthe2previouscase-
controlstudieswithregardtoglucocorticoiduseandrisk
of atrial fibrillation. Van der Hooft et al9analyzed 385
newcasesofatrialfibrillationand6364controlsinacase-
controlstudynestedwithinTheRotterdamStudy,apro-
spective cohort study of persons 55 years and older. Be-
forehand, 50 patients (11%) were excluded because of a
missing index date or because atrial fibrillation was dis-
coveredcoincidently.Currentglucocorticoidusewasas-
Table 3. Association Between Glucocorticoid Use and Atrial Fibrillation or Flutter, Stratified by COPD and Asthma,
Cardiovascular Diseases, and Connective Tissue Disease and RA
Glucocorticoid Use
Patients with COPD and asthma
Never use
Current use
Former use
Patients without COPD and asthma
Never use
Current use
Former use
Patients with cardiovascular diseases
Never use
Current use
Former use
Patients without cardiovascular diseases
Never use
Current use
Former use
Patients with connective tissue disease and RA
Never use
Current use
Former use
Patients without connective tissue disease and RA
Never use
Current use
Former use
No. of Cases/Controls
3847/7692
2288/5045
679/629
880/2018
15465/131674
13702/120199
490/2191
1273/9284
14325/79988
11349/66814
1041/3052
1935/10122
5594/32411
4935/29985
243/359
416/2067
258/299
77/99
101/83
80/107
19402/188275
16280/167798
1010/3580
2112/16897
OR (95% CI)
Unadjusted Adjusteda
1 [Reference]
2.78 (2.20-3.51)
0.91 (0.75-1.10)
1 [Reference]
2.78 (2.14-3.60)
0.86 (0.70-1.06)
1 [Reference]
2.23 (2.05-2.42)
1.22 (1.16-1.29)
1 [Reference]
1.84 (1.68-2.02)
1.03 (0.97-1.09)
1 [Reference]
1.96 (1.72-2.23)
1.15 (1.05-1.26)
1 [Reference]
1.65 (1.43-1.90)
1.03 (0.93-1.13)
1 [Reference]
2.80 (2.58-3.05)
1.22 (1.15-1.30)
1 [Reference]
2.11 (1.92-2.32)
0.99 (0.93-1.06)
1 [Reference]
1.67 (1.08-2.57)
0.90 (0.59-1.38)
1 [Reference]
1.34 (0.83-2.15)
0.88 (0.55-1.40)
1 [Reference]
2.92 (2.72-3.14)
1.29 (1.23-1.36)
1 [Reference]
2.07 (1.91-2.23)
1.01 (0.96-1.06)
Abbreviations: CI, confidence interval; COPD, chronic obstructive pulmonary disease; OR, odds ratio; RA, rheumatoid arthritis.
aAdjusted for a previous diagnosis of cardiovascular diseases, COPD and asthma, diabetes mellitus, renal failure, inflammatory bowel disease, connective tissue
disease and RA, cancer, liver disease and chronic pancreatitis, acute alcohol intoxication, alcoholism-related diseases, hyperthyroidism, and cancer, and use of
nonsteroidal anti-inflammatory drugs or cyclooxygenase 2 inhibitors and cardiovascular, respiratory, and thyroid drugs.
12
2
4
6
10
8
06824 1210
RRCD
OREC
OR = 1.92
OR = 1.79
Figure. Sensitivity analysis that illustrates how strongly an unmeasured
confounder would need to be associated with glucocorticoid use (odds ratio
[OR] for exposure-confounder association, OREC) and atrial fibrillation or
flutter (relative risk of the disease in patients with the confounder, RRCD) to
fully explain our estimates. The solid line represents the adjusted OR and the
dashed line represents the lower limit of the 95% confidence interval for the
adjusted OR.
(REPRINTED) ARCH INTERN MED/VOL 169 (NO. 18), OCT 12, 2009WWW.ARCHINTERNMED.COM
1681
©2009 American Medical Association. All rights reserved.
at Capes Consortia, on October 12, 2009 www.archinternmed.comDownloaded from