Impact of Atrial Fibrillation in Patients With Colorectal Cancer: A National Inpatient Sample Database Analysis

Author Type(s)

Faculty

Document Type

Abstract

Publication Date

10-2021

DOI

10.1093/eurheartj/ehab724.0595

Journal Title

European Heart Journal

Department

Medicine

Abstract

Background

Atrial fibrillation (AF) is the most common cardiac arrhythmia affecting approximately 1–2% overall population (1). Its causal relationship with colorectal cancer (CRC) is much for debate. According to one hypothesis, the presence of autoantibodies directed against ionic channels or acetylcholine receptors can predispose to the development of atrial fibrillation (2–3). Thus, AF may be regarded as an inflammatory complication in patients with colon cancer. Our study objective was to determine if AF impacts the outcome of patients with CRC. Method

We analyzed the National Inpatient Sample (NIS) database from Oct-2015 to Dec 2018 using Stata 16.0. The NIS databases are released under the Healthcare Cost and Utilization Project, which includes inpatient admissions from the United States' participating hospitals. Total population with CRC were identified using their respective ICD-10 diagnostic codes then divided based on AF. To determine atrial fibrillation association with mortality and complications, we used multivariate logistic regression analysis using weights to generate nationally representative results. Later, a propensity-matched population analysis was done for the accuracy of the results. Result

We found 245,305 patients admitted with CRC between Oct 2015 to Dec-2018 in the USA, out of which 28,170 (11.5%) were having AF. The mean age for the patients with AF was 77±10 compare to 65±14 years in those without AF. Patients with AF were associated with higher comorbidities and had a high population percentage with Carlson category three or above. There were 1,456 (5.2%) mortalities in the AF group compared to 5,689 (2.6%) in the other. The higher odds of mortality in patients with AF was present in multivariate logistic regression analysis in both non-propensity matched [1.71 (1.45–2.02), P-value: <0.000] and propensity-matched [1.44 (1.18–1.75), P-value: <0.001] cohorts. Patients with AF were hospitalized longer (9.20±7.8 vs. 6.85±7.0 days), leading to a high admission costs (US$ 25,875±22,875 vs. 20,087±19,314). Odds of complications such as need for blood transfusions [1.61 (1.05–1.29), P-value: 0.005], hemorrhage requiring blood transfusion [1.17 (1.05–1.29), P-value: 0.003], lower-GI bleed [1.31 (1.21–1.43), P-value: <0.001], sepsis [1.45 (1.30–1.62), P-value: <0.001], respiratory failure [1.39 (1.15–1.67), P-value: 0.001] etc. were also higher in group of patients with CRC and AF. Conclusion

In our retrospective, propensity-matched national inpatient sample analyses of patients admitted with colorectal cancer, atrial fibrillation is associated with higher morbidity and mortality. AF was associated with a high burden of complications with prolonged hospital stay leading to increased health care expenditures.

Link to Full Text

Share

COinS