Article, Emergency Medicine

Prevalence of validated risk factors for developing atrial fibrillation—can we identify high-risk ED patients?

Unlabelled imageAmerican Journal of Emergency Medicine (2012) 30, 1581-1587

Brief Report

Prevalence of validated risk factors for developing atrial fibrillationcan we identify high-risk ED patients??

Tyler W. Barrett MD, MSCIa,?, Stephanie A. Couch BSa,

Cathy A. Jenkins MSb, Alan B. Storrow MDa

aDepartment of Emergency Medicine, Vanderbilt University Medical Center, Nashville, TN 37232-4700, USA

bDepartment of Biostatistics, Vanderbilt University School of Medicine, Nashville, TN 37232-2158, USA

Received 1 July 2011; revised 17 August 2011; accepted 18 September 2011

Abstract

Objective: The aim of this study was to investigate whether emergency department (ED) patients who were newly diagnosed with Atrial fibrillation displayed risk factors for incident AF on prior ED visits.

Methods: This was a secondary analysis of a retrospective cohort study of ED patients with symptomatic AF at a tertiary referral center. We selected patients who were newly diagnosed with AF between July 1, 2005, and August 31, 2008, and had at least 1 ED visit before their diagnosis. We calculated the Framingham Heart Study AF risk score for each visit by documenting the presence of the risk factors (age, sex, body mass index, systolic blood pressure, hypertension treatment, PR interval, and ages of clinically significant cardiac murmur and heart failure diagnosis).

Results: Of the 296 patients newly diagnosed with AF, 115 (39%) had at least 1 prior ED visit resulting in 454 ED visits for analysis. The median time from first to last visit was 4 years (interquartile range [IQR], 2.1-5.9). The median age was 66 years (IQR, 49-79 years). Home medications included antihypertensives in 81% of visits, and 60% of visits with available electrocardiograms had a PR interval of 160 milliseconds or more. Heart failure history was reported in 23% of visits. The median AF risk score was 8 (IQR, 4-10) corresponding to a 16% 10-year predicted risk.

Conclusions: Nearly 40% of patients diagnosed with new AF had Previous ED visits and displayed validated risk factors for incident AF. The ED provides an opportunity to identify and educate these patients as well as refer them for primary prevention interventions.

(C) 2012

? Funding Sources: No industry financial support or compensation has been or will be received for conducting this study. Dr Barrett and this study are supported in part by National Institutes of Health grant K23 HL102069 from the National Heart, Lung and Blood Institute. The study was also supported in part by (Vanderbilt CTSA grant UL1 RR024975 from NCRR/NIH) Vanderbilt University School of Medicine Emphasis Program and the Department of Emergency Medicine Research Division.

* Corresponding author. Tel.: +1 615 936 0253; fax: +1 615 936 3754.

E-mail address: [email protected] (T.W. Barrett).

Introduction

Atrial fibrillation , the most common sustained cardiac arrhythmia, affects more than 2.3 million individuals in the United States and accounts for nearly 1% of all emergency department (ED) visits [1,2] The prevalence of AF is expected to more than double by 2050 [1]. Atrial fibrillation is often first diagnosed in the ED when patients present with relevant complaints and receive an electrocardiogram (ECG)

0735-6757/$ – see front matter (C) 2012 doi:10.1016/j.ajem.2011.09.019

or pulse measurements [2]. Atrial fibrillation increases the risk of heart failure and stroke, complications that are commonly treated in the ED. [3] Previous investigations have reported that 1 of every 6 strokes occurs in a patient with AF [4]. Atrial fibrillation increases annual health care costs to approximately 5 times that of patients without AF, with total health care expenditures estimated to cost between $6 and $26 billion [5].

The National Heart, Lung, and Blood Institute, in response to the increasing burden of AF, convened an expert panel that recommended further studies to improve the understanding of the epidemiology, detection, and prevention of AF [3]. The Framingham Heart Study cohort, based on these recommendations, developed and validated a risk score for predicting AF incidence [6,7]. Essential hypertension, heart failure, heart valve dysfunction, diabe- tes, obesity, and PR-interval prolongation are risk factors associated with developing AF [6,7]. We evaluated this score in ED patients with newly diagnosed AF and found that these risk factors are commonly present at the time of diagnosis [8]. Many of these factors are regularly assessed in the ED and could be evaluated quickly without interfering with physicians’ time.

This study’s objective was to investigate whether a cohort of patients who were initially diagnosed with AF in our ED displayed validated AF risk factors on ED visits before their diagnosis. We hypothesized that patients with multiple risk factors for AF might frequent the ED providing a unique opportunity for identifying and educating high-risk in- dividuals. Modifying these factors and initiating appropriate medical therapy might potentially delay or prevent the onset of AF [3].

Materials and methods

Study design

This was a secondary analysis of a retrospective, observational cohort study. We performed a systematic chart review of the Electronic medical records of patients newly diagnosed with AF in the ED. Our medical center’s institutional review board approved the study.

Study setting and population

This study took place at an urban, university-affiliated, adult ED with an annual volume of 55 000 patients. Detailed methodology of the original study has been previously reported [8]. Briefly, adult patients with a primary or supporting International Classification of Diseases, Ninth Edition, ED diagnosis of AF between August 1, 2005, and July 31, 2008, were eligible for inclusion in the original cohort. Inclusion criteria required documented evidence of AF or atrial flutter on an ECG or rhythm strip. We excluded

patients whose ED visits were for complaints unrelated to their AF diagnosis and those with a prior history of AF. Only patients who were newly diagnosed with AF at their ED visit were considered for this investigation. To be eligible for inclusion, patients also had to have EMR evidence of at least 1 ED visit to our institution before their AF diagnosis. Our EMR system includes ED visit documentation beginning in August 2003 providing a 2- to 5-year time interval for recorded ED visits before the patient’s AF diagnosis.

Study protocol

We identified eligible patients and systematically reviewed their EMR for corresponding data. Two abstrac- tors trained on a set of 10 records before data collection. One investigator extensively reviewed all patients’ EMR adhering to strict chart review methodology guidelines [9]. A second investigator randomly selected and reviewed 15% of records to ensure interrater reliability. For each ED visit, we recorded information on patient past and current medical history, home medications, physical examination findings, and diagnostic test results. We abstracted data from both ED and inpatient records. We specifically documented the presence or absence of each of the Framingham risk score’s variables for each visit. The Framingham Heart Study risk score includes the following variables: age, sex, body mass index (BMI), systolic blood pressure (SBP), treatment for hypertension, PR interval, age of clinically significant cardiac murmur diagnosis, and age of heart failure diagnosis [6]. Because of the retrospective nature of the study, some changes from the original Framingham Heart Study were necessary. Rather than age at enrollment as in the Framingham Heart Study, we used the patient’s age at the time of each visit. The information for BMI, SBP, and current treatment for hypertension were also obtained from the time of each visit. We entered data into an electronic spreadsheet (Microsoft Excel 2008, Redmond, Wash).

Data analysis

All analyses were done using the statistical programming language R (version 2.8.1; R Development Core Team 2009, Vienna, Austria). Because subjects have multiple visits, the visit profiles were characterized by the number of visits per person, the length of time between the first and the last (when AF diagnosis made) visit, and the length of time between visits. Continuous measures were summarized using me- dians and interquartile ranges (IQR); categorical variables were summarized with frequencies and percents. We calculated the AF risk score for each subject at each individual ED visit, and the distribution of risk scores in the sample was reported as median (IQR). Missing values for risk factors that comprised the risk score were assigned a value of 0.

Results

Table 2 Descriptive statistics for Framingham risk factors and risk score on ED visits before the visit when AF was newly diagnosed

Framingham risk factor category

BMI

b30

>=30

SBP (mm Hg)

b160

>=160

Treatment for hypertension No

Yes

PR interval (ms)

b160 160-199

>=200

Age of cardiac murmur first diagnosis (y)

<=54

55-64

65-74

75-84

>=85

Age of heart failure first diagnosis (y)

45-54

55-64

65-74

>=75

Framingham risk score, median (IQR)

No. of nonmissing data per ED visit

48

n (%)

37 (77%)

11 (23%)

402

290 (72%)

112 (28%)

442

84 (19%)

358 (81%)

268

108 (40%)

123 (46%)

37 (14%)

140

33 (24%)

35 (25%)

51 (36%)

21 (15%)

176

43 (24%)

35 (20%)

30 (17%)

68 (39%)

8 (4-10)

In the original cohort, there were 834 patients treated for AF-related complaints during the 5-year study period. Of the 834, AF was newly diagnosed in 296 patients (35.4%) with 117 (40%) having documented prior ED visits in our EMR. We excluded 2 additional patients after detailed EMR review revealed a prior diagnosis of AF. The 115 patients who comprised this investigation’s cohort accounted for 569 ED visits between August 1, 2003, and July 31, 2008. This includes the 115 ED visits at which the patients were newly diagnosed with AF and 454 visits before their diagnosis. Women accounted for nearly half of the ED

visits. The median number of ED visits per patient was 3

(IQR, 2-5) with the minimum and maximum number of ED visits per patient of 2 and 23, respectively. The median time interval was 4 years (IQR, 2.1-5.9 years) from a patient’s first reported ED visit to the visit when AF was diagnosed. The median time interval between ED visits within each subject was 221 days (IQR, 95.6-484.8 days). Table 1 reports the baseline characteristics for the 454 ED visits before AF diagnosis.

Table 2 presents the frequency of the Framingham Heart Study AF risk factors present at ED visits before diagnosis.

Home medication regimens included antihypertensives in

81% of ED visits. Electrocardiograms were performed at 268 (59%) of the 454 ED visits before the patients’ AF diagnosis. Of these 268 visits, 160 (60%) ECGs had PR intervals 160 milliseconds or greater including 37 (14%) with evidence of a first-degree atrioventricular block (PR N200 milliseconds). When summarized within an individual, the median risk score was 6.2 (IQR, 2-9.1), corresponding to an 8% 10-year

Table 1 Baseline characteristics per ED visit in a cohort of patients with newly diagnosed AF and ED visits before their diagnosis

No. of nonmissing data per ED visit

Male, n = 234

Female, n = 220

Combined, N = 454

Age, median (IQR)

454

56 (41.3-66.8)

75

(65.0-83.0)

66 (49.0-79.0)

Framingham age category

454

<=49

101 (43%)

14

(6%)

115 (25%)

50-54

11 (5%)

17

(8%)

28 (6%)

55-59

25 (11%)

18

(8%)

43 (9%)

60-64

25 (11%)

5

(2%)

30 (7%)

65-69

26 (11%)

9

(4%)

35 (8%)

70-74

12 (5%)

44

(20%)

56 (12%)

75-79

16 (7%)

27

(12%)

43 (9%)

80-84

8 (3%)

48

(22%)

56 (12%)

>=85

10 (4%)

38

(17%)

48 (11%)

Age risk score, median (IQR)

454

3 (1-5)

6

(3-7)

4 (1-7)

BMI

48

26 (24-30)

SBP (mm Hg)

402

145 (128-164)

PR interval (ms)

268

166 (149-187)

Age of cardiac murmur first diagnosis (y)

140

76 (66-82)

Age of heart failure diagnosis (y)

176

66 (58-79)

Table 3 Calculated AF risk scores for each subject at each individual ED visit before their AF diagnosis

Visit

Case

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

1

4

7

8

2

4

3

11

5

8

8

7

7

7

7

7

7

8

7

8

7

8

7

8

6

11

12

9

9

7

9

9

9

9

9

10

10

8

10

8

11

10

10

11

11

9

8

10

10

9

5

8

8

10

9

9

10

10

10

11

11

-3

12

-2

-1

-1

-1

-1

1

1

13

2

2

14

7

15

10

16

5

5

5

5

5

5

6

6

8

8

8

8

17

14

15

15

18

5

5

6

5

6

5

6

6

8

7

8

8

8

8

19

8

8

7

8

8

9

10

9

10

11

10

9

11

10

10

10

20

7

8

21

5

7

6

6

6

22

-2

-1

0

1

23

13

12

13

13

24

4

3

5

25

7

26

1

1

1

1

27

7

8

28

4

29

10

9

9

10

9

9

30

-3

-3

31

9

11

32

1

2

33

1

1

1

1

1

1

1

1

1

1

1

1

34

4

35

9

9

9

9

10

7

11

8

11

8

8

9

11

8

10

8

11

9

8

8

11

8

36

11

12

11

37

4

4

38

4

6

40

17

18

18

18

17

17

41

1

42

10

9

10

10

10

10

10

10

43

3

3

3

3

3

3

4

4

2

2

2

5

3

3

44

12

13

13

45

9

9

11

11

46

-2

-2

47

2

2

48

10

9

49

1

2

2

3

50

4

51

9

8

52

1

1

53

4

5

54

17

17

17

17

18

18

18

17

17

17

17

17

17

17

18

18

55

7

57

10

11

11

11

10

11

12

10

12

13

13

13

13

11

13

13

12

13

12

58

-1

59

8

8

8

8

Table 3 (continued) Visit

Case 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22

60 7

61 1 1 1 1 3 2 3 3

62 4 4 4

63 8 7 7

64 10 10

65 2

66 6 5 5 6 5 7 6 7 7 8 6

67 5

68 4 4

69 8 9

70 9 9

71 9 9 10 8 11 10

72 4 5

73 1 1 1 1 1

74 9 9

75 3 3

76 1 2

77 3 3

78 1 1

79 18 17 17 20

80 9 10 9 9 9 9 10 9 9 10 11 11 10

81 10 12 10

82 6

83 17 17 18 19

85 4 7 6

86 7

87 6 5 6 8 7 8 8 8 7 8

88 1

89 1 1 1 1 1 1 1 1

90 8 10 9 9 9 9 10

91 8 8 9 8 10 9 9 9 8 9

92 6

93 8

94 1 1

95 2 2

96 9 9 8

97 -2 -1

98 8 8 8

99 1

100 4

101 9

102 10 11 11

103 -2

104 -2 -1

105 8

106 -3 -3 -3

107 6

108 6

111 2

112 2 2

114 5 5

115 8

116 3

117 5 5

118 7 9

(continued on next page)

Table 3 (continued)

Visit

Case

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

121

3

122

10

9

123

10

10

10

126

8

predicted risk. Table 3 reports the AF risk scores for each subject at each individual ED visit before their AF diagnosis.

Limitations

We performed a retrospective chart review, and therefore, this study is subject to the limitations that affect all retrospective cohort analyses. Our cohort consisted only of ED patients with newly diagnosed AF and previous visits to our ED. The inclusion criteria introduced a significant selection bias; however, our objective was only to estimate the frequency of validated risk factors in patients who later developed AF. The retrospective analysis also required that we adjust the definitions for some of the risk factors and assign a zero value for missing data. This very likely might have impacted our estimated Framingham Heart Study risk scores for each visit. Our hospital’s EMR only contains ED visit documentation dating back to 2003, thus limiting the number of prior ED visits available for review.

Discussion

The Framingham Heart Study risk score is a validated prediction rule for identifying individuals at risk for developing AF with greatest risk score weighting assigned to the development of heart failure or significant cardiac murmur before 54 years old [6,7]. We have previously reported that these risk factors are present in many ED patients with newly diagnosed AF [8]. Nearly all of the Framingham risk factors are routinely identified during an ED visit including patient age, sex, SBP, home antihyper- tensive treatment, and PR interval (if an ECG is performed). Body mass index and ages of significant cardiac murmur and heart failure diagnosis might be less reliably documented in the ED record. Body mass index was only documented in 10% of this study’s ED records. In this study, we found that nearly 40% of patients diagnosed with AF in the ED had documented prior visits to the same ED. Furthermore, we identified that these risk factors were evident on prior ED visits and may therefore provide opportunity to identify and educate high-risk patients about their risk for AF, as well as the associated 4- to 5-fold increase risk of stroke, 3-fold increase in developing heart failure, and the potential need to be on lifelong blood thinners [3-5]. Given the significant

complications and health care costs, our goal should be to prevent, or at least delay, the onset of AF. To prevent AF development, it is necessary to identify patients at risk before diagnosis. Our results show that nearly half of the patients who were subsequently diagnosed with AF in the ED presented at high risk (predicted absolute risks of 12% to N30%) for developing AF. Patients reported home antihy- pertensive therapy in more than 80% of the ED visits before AF diagnosis. Emergency department ECGs showed PR intervals 160 milliseconds or greater in 60% of ED visits before AF diagnosis. Of these, 14% had a first-degree atrioventricular block, a conduction disturbance that has been shown to double one’s risk for developing AF [10]. Our EMR review found that both of these risk factors were often ascertained in the ED and therefore could be used to identify patients at risk for developing AF. Emergency department physicians should consider educating these patients about their increased risk for AF and recommending that they follow up with their internist or cardiologist to discuss potential Prevention strategies.

We acknowledge that it is unlikely that identifying these high-risk individuals would have prevented their develop- ment of AF given the relatively short time interval from initial ED visit to AF diagnosis. However, aggressive behavior modification and medical treatments potentially might delay the onset of AF in similar high-risk patients. Research into AF primary prevention has shown benefit of Renin-angiotensin-aldosterone system inhibitors, statin drugs, dietary supplements, and ?-blockers at delaying or prevent- ing incident AF [11-15]. Recent meta-analyses question these treatments’ preventive efficacy, thus further demon- strating the need for additional high-quality clinical trials in at-risk populations [16-17]. The effect of losartan, an angiotensin receptor-blocking drug, on atrial conduction was studied using P-wave duration, and the drug was found to inhibit atrial remodeling, improve atrial conduction disturbance, and suppress AF recurrence [14].

Emergency department physicians are already asked to perform numerous public Health screenings (eg, smoking cessation, immunization status, domestic violence) and education (eg, use of seatbelts and helmets) during our routine patient care. We are not proposing an additional burdensome screening or educational intervention. Instead, we recommend that ED physicians consider a patient’s risk for AF during their standard treatment of patients with hypertension and PR prolongation. A brief discussion with the patient or their

physician might encourage closer surveillance and potentially prevent a devastating complication such as an ischemic stroke in a patient with undiagnosed AF.

Our study suggests that patients often present to the ED and display multiple risk factors for developing AF several years before being diagnosed with AF. Antihypertensive treatment and PR prolongation are easily measurable risk factors that, if present, might warrant a brief discussion about AF and referral to a primary care physician or cardiologist for aggressive primary prevention measures to delay or prevent the onset of AF.

References

  1. Go AS, Hylek EM, Phillips KA, et al. Prevalence of diagnosed atrial fibrillation in adults: national implications for rhythm management and Stroke prevention: the AnTicoagulation and Risk Factors in Atrial Fibrillation (ATRIA) study. JAMA 2001;285(18):2370-5.
  2. McDonald AJ, Pelletier AJ, Ellinor PT, Camargo Jr CA. Increasing US emergency department visit rates and subsequent hospital admissions for atrial fibrillation from 1993 to 2004. Ann Emerg Med 2008;51(1):58-65.
  3. Benjamin EJ, Chen PS, Bild DE, et al. Prevention of atrial fibrillation: report from a national heart, lung, and blood institute workshop. Circulation 2009;119(4):606-18.
  4. Hart RG, Halperin JL. Atrial fibrillation and thromboembolism: a decade of progress in stroke prevention. Ann Intern Med 1999;131: 688-95.
  5. Kim MH, Johnston SS, Chu BC, et al. Estimation of total incremental health care costs in patients with atrial fibrillation in the United States. Circ Cardiovasc Qual Outcomes 2011, doi:10.1161/CIRCOUT- COMES.110.958165 Available at: http://circoutcomes.ahajournals.org.
  6. Schnabel RB, Sullivan LM, Levy D, et al. Development of a risk score for atrial fibrillation (Framingham Heart Study): a community-based cohort study. Lancet 2009;373(9665):739-45.
  7. Schnabel RB, Aspelund T, Li G, Sullivan LM, Suchy-Dicey A, Harris TB, et al. Validation of an atrial fibrillation risk algorithm in whites

and African Americans. Arch Intern Med 2010 Nov 22;170(21): 1909-17.

  1. Barrett TW, Storrow AB, Jenkins CA, Harrell FE, Amdahl J, Russ S, et al. Assessment of the Framingham predictive risk factors among emergency department patients with newly diagnosed atrial fibrilla- tion. Am J Emerg Med 2011 [Epub ahead of print].
  2. Gilbert EH, Lowenstein SR, Koziol-McLain J, Barta DC, Steiner J. Chart reviews in emergency medicine research: where are the methods? Ann Emerg Med 1996;27(3):305-8.
  3. Cheng S, Keyes MJ, Larson MG, et al. Long-term outcomes in individuals with prolonged PR interval or first-degree atrioventricular block. JAMA 2009;301(24):2571-7.
  4. Healey JS, Baranchuk A, Crystal E, et al. Prevention of atrial fibrillation with angiotensin-converting enzyme inhibitors and angio- tensin receptor blockers: a meta-analysis. J Am Coll Cardiol 2005; 45(11):1832-9.
  5. London B, Albert C, Anderson ME, et al. Omega-3 fatty acids and cardiac arrhythmias: prior studies and recommendations for future research: a report from the National Heart, Lung, and Blood Institute and Office Of Dietary Supplements Omega-3 Fatty Acids and their Role in Cardiac Arrhythmogenesis Workshop. Circulation 2007;116(10):e320-5.
  6. Schmieder RE, Julius S, McInnes GT, Zanchetti A, Hua TA, VALUE Trial Group. Reduced incidence of new-onset atrial fibrillation with Angiotensin II receptor blockade: the VALUE trial. J Hypertens 2008;26(3):403-11.
  7. Nomura M, Kawano T, Nakayasu K, Nakaya Y. The effects of losartan on signal averaged P wave in patients with atrial fibrillation. Int J Cardiol 2008;126(1):21-7.
  8. Kulik A, Singh JP, Levin R, Avorn J, Choudhry NK. Association between statin use and the incidence of atrial fibrillation following hospitalization for coronary artery disease. Am J Cardiol 2010;105(12):1655-60.
  9. Schwartz GC, Chaitman BR, Goldberger JJ, Messig M. High-dose atorvastatin and risk of atrial fibrillation in patients with prior stroke or transient ischemic attack: analysis of the Stroke Prevention by Aggressive Reduction in Cholesterol Levels (SPARCL) trial. Am Heart J 2011;161(5):993-9.
  10. Liu T, Korantzopoulos P, Shehata M, Li G, Wang X, Kaul S. Prevention of atrial fibrillation with omega-3 fatty acids: a meta- analysis of randomised clinical trials. Heart 2011;97(3):1034-40.

Leave a Reply

Your email address will not be published. Required fields are marked *