Oncology

Thirty-day mortality in ED patients with new onset atrial fibrillation and actively treated cancer

a b s t r a c t

Objectives: Studies suggest that inflammatory, autonomic, and coagulation alterations associated with cancer may increase incident Atrial fibrillation . New-onset AF is associated with increased mortality in other nonneoplastic disease processes. We investigated the association of active cancer with 30-day mortality in emergency department (ED) patients with new-onset AF.

Methods: We conducted an analysis within an observational cohort study at a tertiary care hospital that included ED patients with new-onset AF. The exposure variable was presence of active cancer. We defined active cancer as the patient received chemotherapy, radiotherapy, or recent cancer- related surgery within 90 days of the ED visit. The primary outcome was 30-day mortality. Logistic regression was used to analyze the association between cancer status and 30-day mortality adjusting for patient age and sex.

Results: During the 5.5-year study period, 420 patients with new-onset AF were included in our cohort, including 37 (8.8%) with active cancer. Patients with active cancer had no clinically relevant differences in their hemodynamic stability. Among the 37 patients with active cancer, 9 (24%) died within 30 days. Of the 383 patients without active cancer, 11 (3%) died within 30 days. After adjusting for age and sex, ac- tive cancer was an independent predictor of 30-day mortality, with an adjusted odds ratio of 10.8 (95% confi- dence interval, 3.8-31.1).

Conclusions: Among ED patients with new-onset AF, active cancer appears to be associated with 11-fold in- creased odds of 30-day mortality; new-onset AF may represent progressive organ dysfunction leading to an in- creased risk of short-term mortality in patients with cancer.

(C) 2015

? Conflicts of Interest/Disclosures: There are no conflicts of interest in connection with this submission, or are there any copyright constraints. No industry financial support or compensation has been or will be received for conducting this study. Dr Barrett serves as a scientific consultant for Red Bull GmbH, Fuschl am See, Salzburg (Austria), and Boehringer Ingelheim Pharmaceuticals, Inc, Ridgefield, CT, and has received research funding from Alere, Boehringer Ingelheim, and Janssen. Dr Barrett is a member of the American Heart Association’s The Guideline Advantage Research Subcommittee and serves on the Editorial Board for Annals of Emergency Medicine. Dr Self is a paid advisor for BioFire Diagnostics, Inc, and Venaxis, Inc, and has received research funding from CareFusion, BioMerieux, Affinium Pharmaceuticals, Astute Medical, BRAHMS GmbH, Pfizer, Rapid Pathogen Screening, Venaxis, and BioAegis. Dr Lardaro has no financial dec- larations and no potential conflicts of interest.

?? Funding Sources: Dr Barrett and this study are funded by NIH Grant K23 HL102069

from the National Heart, Lung and Blood Institute, Bethesda, MD. Dr Self is supported by NIH Grant K23GM110469 from the National Institute of General Medical Sciences. This study was also supported by Grant UL1 TR000445 from the National Center for Advancing Translational Sciences/NIH. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.

? Role of the Sponsors: The funding organizations had no role in the design and conduct of

the study; the collection, analysis, and interpretation of the data; the preparation, review, or approval of the manuscript; or the decision to submit the manuscript for publication.

* Corresponding author at: Department of Emergency Medicine, Vanderbilt University School of Medicine, 703 Oxford House, 1313 21st Ave South, Nashville, TN 37232-4700. Tel.: +1 615 936 0253; fax: +1 615 936 1316.

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

  1. Introduction

In the United States, between 3 and 6 million individuals have Atrial fibrillation and the prevalence is expected to double by 2050 [1-3]. Among individuals 40 years and older, the lifetime risk for developing AF is estimated to be 1 in 4 [3-5]. Atrial fibrilla- tion is associated with many adverse sequelae, including heart fail- ure [6], a 5-fold increase in the risk of cerebrovascular accident representing an estimated 4 to 13 percent annual risk [4,6,7], and a nearly 2-fold increase in mortality [6,8,9]. Atrial fibrillation has been found to occur more frequently in patients with cancer [10,11]. Studies have suggested that the increased inflammation, autonomic, and coagulation alterations associated with cancer may increase patient risk for developing AF [10]. New-onset AF is associated with increased mortality in patients with other disease processes such as severe sepsis and chronic kidney disease [9,12]. Hu et al [11], using a large Taiwanese population database and In- ternational Classification of Diseases, Ninth Revision code definitions for exposures and outcomes, previously reported that new-onset AF did not impact long-term mortality among cancer patients. Al- though AF increases an individual’s lifetime risk of death and stroke

http://dx.doi.org/10.1016/j.ajem.2015.07.033

0735-6757/(C) 2015

[4,6,7], the 30-day risk following an emergency department (ED) evaluation for AF is relatively low with a combined incidence of 1% to 3% [13-16]. We have previously identified several factors that are associated with 30-day adverse events in patients with AF [14,15]. To our knowledge, no prior study has evaluated whether a preexisting diagnosis of active cancer is associated with increased short-term mortality among patients diagnosed as having new-onset AF. We investigated the association of active cancer with 30-day mortality in ED patients with new-onset AF.

  1. Methods

We conducted an analysis within an observational cohort study. The cohort combined 2 databases of patients who presented to our adult ED for evaluation of symptomatic AF. The Risk Estimator decision aid for Atrial Fibrillation (RED-AF) database was a retrospective cohort of adult ED patients presenting to our center for symptomatic AF or atrial flutter between August 1, 2005, and July 31, 2008 [14]. The Atrial Fibril- lation and Flutter Outcomes and Risk Determination (AFFORD) database

Table 1

Baseline characteristics of the cohort (n = 420)

Variable

Active cancer (n = 37)

No active cancer (n = 383)

Age (y)

67 (62, 78)

64 (52, 76)

Female

7 (19%)

145 (38%)

White, non-Hispanic

32 (87%)

322 (84%)

Black

4 (11%)

53 (14%)

White, Latino or Hispanic

0

3 (1%)

Asian

0

1 (0.3%)

Native American

0

1 (0.3%)

History of cancer

37 (100%)

63 (16%)

Chemotherapy within 90 d of diagnosis of AF

26 (70%)

0

Radiation therapy within 90 d of diagnosis of AF

9 (24%)

0

Cancer related surgery within 90 d of diagnosis of AF

8 (22%)

0

Type of cancer

Bladder

2 (5%)

0

Brain

0

1 (0.3%)

Breast

1 (3%)

7 (2%)

Colorectal

3 (8%)

1 (0.3%)

Gynecologic

1 (3%)

6 (2%)

Liver

1 (3%)

1 (0.3%)

Lung

10 (27%)

3 (1%)

Lymphoma

4 (11%)

4 (1%)

Melanoma

5 (14%)

0

Oropharyngeal

1 (3%)

1 (0.3%)

Pancreas

0

1 (0.3%)

Prostate

3 (8%)

7 (2%)

Skin

0

13 (3%)

Thyroid

0

4 (1%)

Height (meter)

1.8 (1.7, 1.8)

1.8 (1.7, 1.8)

Weight (kg)

78 (69, 92)

84 (71, 100)

BMI

25.4 (22.8, 28.4)

27.5 (23.8, 32.2)

Stroke/Transient ischemic attack

3 (8%)

49 (13%)

Coronary artery disease

8 (22%)

92 (24%)

Heart failure

2 (5%)

55 (14%)

Hypertension

22 (60%)

236 (62%)

Chronic obstructive pulmonary disease

6 (16%)

45 (12%)

Vascular disease

4 (11%)

37 (10%)

Diabetes mellitus

7 (19%)

82 (21%)

CHA2DS2-VASc

3 (1, 4)

3 (1, 4)

Warfarin use

2 (5%)

16 (4%)

?-Blocker use home medication

11 (30%)

121 (32%)

Calcium-channel blocker home medication

3 (8%)

44 (12%)

Statin home medication

13 (35%)

116 (30%)

Smoker, Current

2 (5%)

65 (17%)

Smoker, Former

10 (27%)

39 (10%)

Triage ventricular rate at index ED visit (beats/min)

120 (108, 154)

118 (96, 142)

Systolic blood pressure at index ED visit (mm Hg)

126 (101, 141)

134 (119, 154)

Total white blood cell count (x103/uL) on index ED visit

9.0 (6.1, 16.2)

8.2 (6.7, 10.7)

Hematocrit (%) on index ED visit

35 (31, 40)

42 (38, 45)

Sodium (mmol/L) on index ED visit

137 (132, 140)

139 (136, 141)

Potassium (mmol/L) on index ED visit

4.3 (3.8, 4.5)

3.9 (3.6, 4.3)

Blood urea nitrogen (mg/dL) on index ED visit

20 (13.5, 31.5)

17 (13, 24)

Creatinine (mg/dL) on index ED visit

1.1 (1.0, 1.5)

1.1 (0.9, 1.3)

Admitted to floor bed on index ED visit

30 (81%)

299 (78%)

Admitted to intensive care unit bed on index ED visit

3 (8%)

24 (6%)

Diagnosed with sepsis on index ED visit

0

2 (b 1%)

Anemia listed as a contributing diagnosis on index ED visit

1 (2.7%)

5 (1.3%)

Discharged home on index ED visit

4 (11%)

60 (16%)

palliative care/hospice consultation within 30 d of index ED visit

11 (30%)

16 (4%)

Median (IQR) days until death

17 (12, 23)

6 (4, 11)

Death within 30 d

9 (24%)

11 (3%)

Categorical data are presented as number of nonmissing values (percentage). Continuous variable data are presented as median (IQR). IQR, interquartile range; CHA2DS2-VASc, Congestive heart failure (1 point), Hypertension (1 point), Age 75 or older (2 points), Diabetes (1 point), Previous Stroke or transient ischemic attack (2 points), Vascular Disease (1 point), Age 65-74 y (1 point), and Sex (Female 1 point).

Table 2

Multivariable model for predicting 30-day mortality among patients with new-on- set AF

Variable

Adjusted odds ratio

95%

Confidence interval

Active cancer

10.8

3.8-31.1

Age (per year)

1.1

1.0-1.1

Male

1.2

0.4-3.4

enrolled a prospective cohort of ED patients with symptomatic AF and atrial flutter from June 8, 2010, to February 28, 2013 [15]. The details of both studies have been previously reported [14,15,17]. The study hos- pital is an urban, university-affiliated, tertiary care referral center that treats 70000 adult patients annually. We queried the databases and se- lected all patients with newly diagnosed AF on their index ED evalua- tion. The primary exposure variable was presence of active cancer at the time of AF diagnosis in the ED. We documented whether each patient had a history of cancer, active cancer currently under- going treatment, or no documented cancer. We defined active can- cer as the patient having received chemotherapy, radiotherapy, or cancer-related surgery within 90 days of the ED visit when AF was newly diagnosed. The primary outcome was 30-day mortality ob- tained through telephone interviews and medical record review. Two investigators (T.L. and T.W.B.) systematically reviewed all re- cords to determine cancer classification with the final classification decided by consensus. The primary outcome was death within 30 days of the index ED presentation. We reviewed medical records and recorded the physician reported cause of death when available. Our hospital’s institutional review board reviewed and approved this study.

We calculated descriptive statistics on the baseline characteris-

tics for the cohort stratified by active cancer classification. We an- alyzed the association between active cancer at the time of new AF diagnosis and 30-day mortality using multivariable logistic regres- sion. We included age and sex as covariates in the regression model. All analyses were done with STATA (StataCorp, College Sta- tion, TX).

  1. Results

During the 5.5 years of ED evaluations, 420 patients with new- onset AF were enrolled, including 37 (9%) with active cancer (Figure A.1). Table 1 presents the baseline characteristics. Nine (24%) of the 37 patients with active cancer and 11 (3%) of the 383 patients without active cancer died within 30 days of their index ED evaluation. In the multivariable logistic regression model, active cancer (adjusted odds ratio, 10.8; 95% confidence interval, 3.8-31.1) was significantly associated with 30-day mortality after adjusting for patient age and sex (Table 2). Of the 9 patients with active can- cer who died within 30 days, 7 of the deaths were directly related to their cancer. Meanwhile, most deaths in the group without active cancer were attributed to AF-related complications, including stroke and decompensated heart failure. A detailed list of the cause of death for all 20 patients is provided in Table A.1. The median (interquartile range) days until death among the patients with and

without active cancer were 17 (12, 23) days and 6 (4, 11) days, respec- tively. All 20 patients were hospitalized after their index ED visits in- cluding 3 patients without active cancer who were admitted to the intensive care unit.

  1. Discussion

To our knowledge, this is the first study to investigate wheth- er a preexisting diagnosis of active cancer is associated with in- creased 30-day mortality among patients diagnosed as having new-onset AF. This investigation found an 11-fold increase in the odds of death within 30-days of a new AF diagnosis among individuals with active cancer compared with those without ac- tive cancer. Atrial fibrillation is known to have an inflammatory component to its development, similar to other Disease states, such as sepsis and postoperative period [9,10,18,19]. In this study, we found active cancer to be a major risk factor for short-term mortality among patients with new-onset AF. We hypothesize that new AF may be an indicator that a patient’s cancer has progressed, with increased systemic inflammation triggering the AF. Patients with active cancer had no clinically relevant differences in their hemodynamic stability. A similar percentage of patients with and without active cancer were hospitalized with most admitted to a floor bed. Most patients in this cohort had hypertension, nearly 1 in 4 had documented coronary artery disease, and 14% had heart failure at the time of their Initial AF diagnosis. Therefore, these patients had established risk factors for developing AF independent of their cancer status [5,18].

Patients with cancer that is being actively treated are clearly at

increased risk for developing opportunistic infection related to chemotherapy-induced neutropenia, indwelling central access catheters, and frequent Hospital visits [19]. Cancer treatments also have known adverse effects that might increase an individual’s risk of death [20]. Patients with active cancer died due to complica- tions of their cancer, whereas individuals without active cancer more often died due to AF-related complications. Nearly 1 in 6 of the patients without active cancer had a documented remote histo- ry of cancer and may also have been exposed to similar treatments during their lives.

Our conclusions must be interpreted in the context that only

20 patients experienced our primary outcome. Therefore, we could only adjust for a limited number of potential confounders in our multivariable analysis. The study should not be used to prognosticate an individual’s 30-day mortality. All of the patients who died within 30 days were hospitalized after their index ED evaluation. There is the potential that hospitalization alone in- creases the potential mortality for patients with active cancer. This study is an observational cohort analysis and is subject to the inherent limitations of such study designs. There is the poten- tial that patients with an undiagnosed malignancy may have been misclassified as not having active cancer at the time of their AF diagnosis. We did review medical records both at least 90 days before and after their ED evaluation to identify any documenta- tion of cancer and related treatment. This study was performed at a single, academic, tertiary care referral center that has a large oncology patient population, and our results are limited by the potential selection bias associated with our high-acuity pa- tient population. Acknowledging these limitations, our results suggest that new-onset AF may represent progressive organ dys- function leading to an increased risk of short-term mortality in patients with cancer.

Appendix

Table A.1

Baseline characteristics and causes of death for 20 patients who died within 30 days of index ED visit

Age (y)

Sex

Active cancer

Type of cancer

Days until death

Cause of death

Palliative care

CHA2DS2-VASc

Total white count (x103/uL) at index ED visit

Hematocrit (%) at index ED visit

Creatinine (mg/dL) at index ED visit

Troponin (ng/mL) at index ED visit

70

Male

Yes

Prostate

17

Chronic kidney disease

Yes

2

8.9

37

6.3

0.15

72

Male

Yes

Melanoma

17

Cancer related

No

3

4.3

31

1.3

b0.01

83

Male

Yes

Colon

21

Cancer related

Yes

3

6.6

32

2

b0.01

58

Male

Yes

Lung

41

Cancer related

No

0

57.5

31

1.1

0.74

78

Male

Yes

Bladder

20

Cancer related

No

4

12.4

26

1.2

0.23

67

Male

Yes

Head and neck

24

Cancer related

Yes

4

15.4

39

1.2

1.1

81

Male

Yes

Prostate

14

Cancer related

Yes

6

4.9

30

1

0.02

69

Female

Yes

Breast

9

Sepsis

No

3

7.2

46

2.9

0.05

82

Male

Yes

Lung

8

Cancer related

Yes

6

1.4

32

1.1

0.05

91

Male

No

History of prostate

9

Stroke

No

6

13.7

38

1.3

0.02

79

Male

No

History of lung

4

Sepsis

Yes

6

8.4

37

1.9

b0.01

69

Female

No

11

Stroke

No

5

7.6

33

0.9

0.02

57

Female

No

History of breast

8

COPD

No

3

11.8

43

0.7

0.03

76

Female

No

30

Stroke

Yes

5

5.9

21

2.2

0.07

85

Female

No

6

Stroke

Yes

4

9.2

29

1

0.01

82

Male

No

6

Sepsis

Yes

7

15.5

27

1.9

0.14

75

93

Female

Female

No

No

New diagnosis of lung on index ED admission

5

4

Cerebral edema from new brain metastases

Heart failure

Yes

Yes

5

3

8.9

5.1

43

36

0.8

0.9

b0.01

0.2

65

Male

No

2

Aortic aneurysm

No

5

22.8

29

1.8

Not collected

85

Male

No

28

Heart failure

Yes

7

19.7

40

1.6

0.07

COPD, chronic obstructive pulmonary disease; CHA2DS2-VASc, Congestive heart failure (1 point), Hypertension (1 point), Age 75 or older (2 points), Diabetes (1 point), Previous Stroke or transient ischemic attack (2 points), Vascular Disease (1 point), Age 65-74 y (1 point), and Sex (Female 1 point).

Figure A.1. Study flow diagram. This figure shows the development of the cohort through the merger of the RED-AF and AFFORD databases.

References

  1. Camm AJ, Lip GY, De Caterina R, Savelieva I, Atar D, Hohnloser SH, et al. Committee for Practice Guidelines. 2012 focused update of the ESC Guidelines for the manage- ment of atrial fibrillation: an update of the 2010 ESC Guidelines for the management of atrial fibrillation. Developed with the special contribution of the European Heart Rhythm Association. Eur Heart J 2012;33:2719-47.
  2. Go AS, Mozaffarian D, Roger VL, Chang Y, Henault LE, Selby JV, et al. Prevalence of di- agnosed atrial fibrillation in adults: national implications for rhythm management and Stroke prevention: the AnTicoagulation and Risk Factors in Atrial Fibrillation (ATRIA) Study. JAMA 2001;285:2370-5.
  3. Go AS, Mozaffarian D, Roger VL, Benjamin EJ, Berry JD, Blaha MJ, et al. American Heart Association Statistics, Committee, Stroke Statistics, Subcommittee. Heart dis- ease and stroke statistics-2014 update: a report from the American Heart Associa- tion. Circulation 2014;129:e28-292.
  4. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham Study. Stroke 1991;22:983-8.
  5. Lloyd-Jones DM, Wang TJ, Leip EP, Larson MG, Levy D, Vasan RS, et al. Lifetime risk for development of atrial fibrillation: the Framingham Heart Study. Circulation 2004;110:1042-6.
  6. Stewart S, Hart CL, Hole DJ, McMurray JJ. A population-based study of the long-term risks associated with atrial fibrillation: 20-year follow-up of the Renfrew/Paisley study. Am J Med 2002;113:359-64.
  7. Hart RG, Pearce LA, Aguilar MI. Meta-analysis: antithrombotic therapy to prevent stroke in patients who have nonvalvular atrial fibrillation. Ann Intern Med 2007; 146:857-67.
  8. Benjamin EJ, Wolf PA, D’Agostino RB, Silbershatz H, Kannel WB, Levy D. Impact of atrial fibrillation on the risk of death: the Framingham Heart Study. Circulation 1998;98:946-52.
  9. Walkey AJ, Wiener RS, Ghobrial JM, Curtis LH, Benjamin EJ. Incident stroke and mor- tality associated with new-onset atrial fibrillation in patients hospitalized with se- vere sepsis. JAMA 2011;306:2248-54.
  10. Farmakis D, Parissis J, Filippatos G. Insights into onco-cardiology: atrial fibrillation in cancer. J Am Coll Cardiol 2014;63:945-53.
  11. Hu YF, Liu CJ, Chang PM, Tsao HM, Lin YJ, Chang SL, et al. Incident thromboembolism and heart failure associated with new-onset atrial fibrillation in cancer patients. Int J Cardiol 2013;165:355-7.
  12. Reinecke H, Brand E, Mesters R, Schabitz WR, Fisher M, Pavenstadt H, et al. Di- lemmas in the management of atrial fibrillation in chronic kidney disease. J Am Soc Nephrol 2009;20:705-11.
  13. Atzema CL, Austin PC, Chong AS, Dorian P. Factors associated with 90-day death after emergency department discharge for atrial fibrillation. Ann Emerg Med 2013;61 [539-548 e531].
  14. Barrett TW, Martin AR, Storrow AB, Jenkins CA, Harrell Jr FE, Russ S, et al. A clinical prediction model to estimate risk for 30-day adverse events in emergency depart- ment patients with symptomatic atrial fibrillation. Ann Emerg Med 2011;57:1-12.
  15. Barrett TW, Storrow AB, Jenkins CA, Abraham RL, Liu D, Miller KF, et al. The AFFORD Clinical Decision Aid to Identify Emergency Department Patients With Atrial Fibrilla- tion at Low Risk for 30-Day Adverse Events. Am J Cardiol 2015;115:763-70.
  16. Scheuermeyer FX, Grafstein E, Stenstrom R, Innes G, Heslop C, MacPhee J, et al. Thir- ty-day and 1-year outcomes of emergency department patients with atrial fibrilla- tion and no acute underlying medical cause. Ann Emerg Med 2012;60:755-65.
  17. Barrett TW, Storrow AB, Jenkins CA, Harrell Jr FE, Miller KF, Moser KM, et al. Atrial fibrillation and flutter outcomes and risk determination (AFFORD): design and ratio- nale. J Cardiol 2011;58:124-30.
  18. January CT, Wann S, Alpert JS, Calkins H, Cleveland JC, Cigarroa JE, et al. 2014 AHA/ ACC/HRS Guideline for the Management of Patients With Atrial Fibrillation: A Re- port of the American College of Cardiology/American Heart Association Task Force

on Practice Guidelines and the Heart Rhythm Society. J Am Coll Cardiol 2014;64: e1-76.

  1. Caggiano V, Weiss RV, Rickert TS, Linde-Zwirble WT. Incidence, cost, and mortality of neutropenia hospitalization associated with chemotherapy. Cancer 2005;103:1916-24.
  2. Kuderer NM, Dale DC, Crawford J, Cosler LE, Lyman GH. Mortality, morbidity, and cost associated with febrile neutropenia in adult cancer patients. Cancer 2006;106:2258-66.