Abstract
Objective
We aimed to determine the incidence of new onset atrial fibrillation (NOAF) in a cohort of intensive care unit (ICU) patients and, further, identify commonly utilised pharmacological strategies for its management in patients with and without sepsis.
Design
A multicentre, retrospective observational study was conducted.
Setting
Twelve ICUs in Queensland, Australia.
Participants
Adult patients, excluding those with cardiothoracic surgical diagnoses, admitted to a participating ICU from 2015 to 2021.
Main outcome measures
Main outcome measures included the incidence of NOAF in ICU, association of NOAF with illness severity and outcomes, cardiac rhythm at ICU discharge, and incidence of pharmacological intervention for NOAF in the ICU.
Results
NOAF occurred in 8.4 % of included ICU admissions, and was associated with higher illness severity, length of stay, and mortality. The majority of patients who experienced NOAF and survived their ICU stay were discharged from the ICU in a sinus rhythm (68.6 %). Patients with sepsis-associated NOAF were more likely to be in a sinus rhythm at ICU discharge than patients with NOAF without sepsis (72.2 vs 65.7 %). Amiodarone was frequently (50.4 %) prescribed to patients both with (56.5 %) and without (45.3 %) sepsis.
Conclusion
NOAF was common amongst patients admitted to the ICU, and amiodarone is commonly prescribed. Future studies are required to determine the optimal short- and long-term management strategies for NOAF complicating critical illness.
Keywords: New onset atrial fibrillation, Intensive care unit, Sepsis, Amiodarone
1. Introduction
New onset atrial fibrillation (NOAF) is defined as atrial fibrillation (AF) that occurs in the absence of a known history of AF; often in association with a precipitating and potentially reversible factor such as illness.1 NOAF has been previously shown to occur in 5 to 15 % of patients admitted to the intensive care unit (ICU),2 and in up to half of patients with septic shock.3
Previous studies have demonstrated that patients who develop NOAF during their ICU admission are at a higher risk of both in- and post-hospital death,[4], [5], [6] and have an increased risk of readmission to hospital with AF,6 heart failure,6 and thromboembolism[4], [5], [6] than those who do not develop NOAF. However, whether there exists a causal association between critical illness-associated NOAF and poor outcomes (cf. an epiphenomenon) is less clear.7
There is uncertainty regarding how to best manage NOAF in critically ill patients, and there is significant variability in the reported efficacy of numerous commonly employed pharmacological interventions.[8], [9], [10], [11], [12] Specific arrhythmogenic precipitants in the context of critical illness (e.g. systemic inflammation, increased sympathetic drive, exogenous catecholamine exposure, volume shifts) may mean that the adoption of guidelines developed for non-critically ill patients with AF does not apply in the ICU context.13,14
We aimed to define the epidemiology of NOAF in the Australian ICU setting and, additionally, aimed to assess the frequency of common pharmacological interventions used to treat NOAF in the ICU. Furthermore, we aimed to compare the epidemiology and pharmacological management of NOAF in ICU patients with and without sepsis.
2. Methods
2.1. Study design and setting
A multicentre, retrospective cohort study was conducted in 12 closed model, mixed (medical and surgical) ICUs in Queensland, Australia, using routinely collected, electronic medical record –based clinical data.
2.2. Study sites and patient identification
Data were collected from five tertiary, three outer metropolitan, and four regional ICUs. All adult (≥18 years of age) patients admitted between 1 January 2015 and 31 December 2021 were eligible for analysis. Patients were excluded from the study if they had a rhythm of AF at the time of ICU admission, were admitted solely for palliation or organ donation, were admitted from another ICU, did not have any hourly observations recorded or were missing all data relating to their cardiac rhythm. Cardiothoracic surgical patients were excluded given this population likely have distinct postoperative triggers for NOAF.15 Data related to a patient’s first admission to an ICU during the study period were used with all data arising from re-admissions excluded.
2.3. Data sources
Data were collected from the electronic medical record of all centres using the eCritical MetaVision™ (iMDsoft, Boston, MA, USA) clinical information systems with registry data from the Australian and New Zealand Intensive Care Society Centre for Outcome and Resource Evaluation Adult Patient Database (ANZICS CORE APD). The complete dataset includes patient demographics, diagnoses, severity of illness, outcomes, hourly haemodynamic measurements, and administered medications for each 24-h period in the ICU. Heart rhythm data are included in the hourly nursing observations and are inputted and validated by the clinical staff caring for the patients as part of routine clinical practice. Primary and secondary diagnoses were obtained from the International Classification of Diseases 10 Australia Modification (ICD-10-AM) codes. Mortality data were collected from the ANZICS CORE APD and Queensland Birth, Death, and Marriage Registry. Some post ICU discharge deaths occurring outside Queensland may not have been captured. Admission diagnoses were categorised to optimise data accuracy and interpretability (Supplementary Methods, Table S1).
2.4. Definitions
NOAF was defined as AF during an ICU admission that was not present on admission to the ICU.16 Patients with AF-related ICD-10-AM codes were not specifically excluded from the primary analysis due to the difficulty in ascribing these diagnoses as pre-existing or relating to the current admission. A discrete episode of AF was defined as having another rhythm recorded for at least the hour before and after. Rapid ventricular rate is variably defined in the literature and guidelines,7 but for the purposes of this study, it was defined as a rate greater than 100 beats per minute.9 The last recorded cardiac rhythm in ICU was used as the rhythm at time of discharge, while persistent reversion to sinus rhythm was defined as the absence of AF for at least 24 h prior to ICU discharge. A medication was considered to be administered after NOAF onset if it was given on the same day as NOAF onset or later. Administration of a medication within a day of discharge was considered to be ongoing at discharge from ICU. The CHA2DS2-VASC score,17 Charlson-defined comorbidities, and index were calculated from ICD-10 codes18,19 (Supplementary Methods, Tables S2 and S3). Sepsis and septic shock were defined using the Sepsis-3 criteria utilising previously published methodology.[20], [21], [22]
2.5. Statistical analysis
An initial descriptive analysis was conducted with continuous data presented as median and interquartile range (IQR) and discrete data presented as count and percentage. Between group comparisons were made using the Mann–Whitney U test for continuous data, Fisher’s exact test for binary categorical data and the chi-square test for categorical data with multiple categories. No adjustment for multiple comparisons was conducted due to the exploratory and descriptive nature of this study.
Regression models were constructed to explore factors contributing to the development of NOAF in ICU as well as its relationship to mortality. Covariates for each model were selected a priori based on their availability within the dataset and relevance to the dependent variable.
A generalised mixed effects regression was constructed to assess for factors that contributed to the development of NOAF. ICU site ID was included as a random effect. Model performance was assessed by Variance Inflation Factor (VIF) for multicollinearity, partial regression plots, the Hosmer–Lemeshow goodness-of-fit test, calibration plot, as well as using the area under the receiver operating characteristic curve.
A stratified, time-dependent Cox proportional hazards model with mixed effects was developed to assess the impact of NOAF on 1-year survival following ICU admission. Time to event was defined as the number of days from ICU admission until death or censoring at 365 days. To account for the risk of immortal time bias, NOAF and fulfilment of Sepsis-3 criteria were treated as time-dependent variables, with patients being treated as exposed from their respective onset times. The APACHE III score and Charlson comorbidity index were categorised into quartiles and used for stratifications due to being non-proportional hazards. Clustering via ICU site was conducted to account for site-level confounding. For the purpose of visualisation, adjusted survival curves were generated using calculated marginal estimates weighted by the empirical distribution of the covariates. Bootstrap resampling with replacement of the entire procedure was conducted for 1000 iterations to estimate 95 % confidence intervals (CI) of the adjusted survival curves. Unadjusted Kaplan–Meier curves were also plotted for comparison.
Missing data were not imputed—the number of present data points for each variable is included in the descriptive analysis and all regression analyses were conducted using complete cases only. All analyses were conducted using Python 3.11 (Python Software Foundation, https://www.python.org/) and R 4.5.1 (R Core Team, 2025).
2.6. Ethical considerations
A waiver of consent was granted by the Metro South Hospital and Health Service Human Research Ethics Committee (HREC/2022/QMS/82024).
3. Results
During the study period, there were 89,184 admissions for 74,522 patients, for which the first admission for 46,599 patients met the pre-specified inclusion criteria. NOAF occurred in 8.4 % (n = 3911/46,599) of included patients (see Fig. 1).
Fig. 1.
Flowchart detailing enrolment, exclusions, and patients included with and without new onset atrial fibrillation (NOAF).
Baseline characteristics of patients, with and without NOAF, are outlined in Table 1. Patients with NOAF were more likely to be male, over 60 years of age, have a higher Charlson comorbidity index, and a higher incidence of a documented history of ischaemic heart disease and/or congestive heart failure. Patients with NOAF complicating their ICU admission were also more likely to have both sepsis (45.3 % vs 25.2 %, p < 0.001) and septic shock (23.3 % vs 7.4 %, p < 0.001) than patients who did not have NOAF. Markers of illness severity were also greater for patients with NOAF than those without, including APACHE III score (61 vs 48, p < 0.001), Australian and New Zealand Risk of Death score (6.5 vs 1.9, p < 0.001), and admission Sequential Organ Failure Assessment score (6 vs 4, p < 0.001). Patients with NOAF additionally had a higher lactate on the day of admission (2.4 mmol/L vs 1.8 mmol/L, p < 0.001).
Table 1.
Admission characteristics of intensive care unit (ICU) patients with new onset atrial fibrillation (AF) and without AF.
| Overall cohort | |||
|---|---|---|---|
| Incidence of new onset AF during ICU (%) | 3911/46,599 (8.4 %) |
||
| Without AF (n = 42688) | New onset AF (n = 3911) | p | |
| Demographics | |||
| Sex (% female) | 17,975 (42.1 %) | 1501 (38.4 %) | <0.001 |
| Age (median [IQR]) | 58.0 [43.0–69.0] | 65.0 [52.0–74.0] | <0.001 |
| Over 60 years old (%) | 19,067 (44.7 %) | 2337 (59.8 %) | <0.001 |
| Charlson comorbidity index (median [IQR]) | 3.0 [1.0–5.0] | 4.0 [2.0–6.0] | <0.001 |
| History of ischaemic heart disease | 1982 (4.6 %) | 386 (9.9 %) | <0.001 |
| History of congestive heart failure | 3550 (8.3 %) | 660 (16.9 %) | <0.001 |
| Admission characteristics | |||
| APACHE diagnosis group | <0.001 | ||
| Cardiovascular | 5849 (13.7 %) | 815 (20.8 %) | |
| Neurological | 7638 (17.9 %) | 454 (11.6 %) | |
| Respiratory | 6023 (14.1 %) | 524 (13.4 %) | |
| Gastrointestinal | 7478 (17.5 %) | 638 (16.3 %) | |
| Genitourinary | 2218 (5.2 %) | 148 (3.8 %) | |
| Haematological | 110 (0.3 %) | 16 (0.4 %) | |
| Metabolic | 3864 (9.1 %) | 265 (6.8 %) | |
| Trauma | 4744 (11.1 %) | 383 (9.8 %) | |
| Sepsis | 3386 (7.9 %) | 546 (14.0 %) | |
| Other | 1378 (3.2 %) | 122 (3.1 %) | |
| Sepsis 3.0 criteria (% yes) | 10,748 (25.2 %) | 1771 (45.3 %) | <0.001 |
| Septic shock (% yes) | 3167 (7.4 %) | 913 (23.3 %) | <0.001 |
| Emergency ICU admission | 29,053 (68.1 %) | 3213 (82.2 %) | <0.001 |
| Surgical admission diagnosis (%) | 22,093 (51.8 %) | 1554 (39.7 %) | <0.001 |
| Markers of illness severity | |||
| APACHE III score | 48 [34.0–66.0] | 61 [44.0–83.0] | <0.001 |
| ANZROD (% ROD) | 1.9 [0.6–8.2] | 6.5 [1.4–26.4] | <0.001 |
| Admission SOFA score | 4 [2.0–6.0] | 6 [4.0–9.0] | <0.001 |
| Admission SOFA score cccomponent | 0 [0.0–3.0] | 3 [0.0–3.0] | <0.001 |
| Highest lactate day 1 (mmol/L) | 1.8 [1.2–3.0] | 2.4 [1.5–4.5] | <0.001 |
ANZROD: Australian and New Zealand Risk of Death model; APACHE III: Acute Physiology and Chronic Health Evaluation III; IQR: interquartile range; SOFA: Sequential Organ Failure Assessment score; ROD: risk of death.
Characteristics of AF in patients with NOAF are reported in Table 2. Of those patients who had NOAF during their ICU admission, the median time from admission to onset of AF was 33 h (IQR: 12–63) and the median number of episodes of AF was 2 (IQR: 1–3). The median duration of an episode of AF was 2 h (IQR: 1–8). The median maximum heart rate of all patients with NOAF was 122 beats per minute (IQR: 97–142) and just over half (51 %) of patients with NOAF were in a rapid ventricular rate at the time of AF onset. Only 4.3 % of patients had a new vasoactive requirement associated with the onset of AF. Over two thirds of patients who had NOAF during their ICU admission, and who survived their ICU admission, were discharged from the ICU in sinus rhythm (68.6 %) and 53.2 % had no further AF for at least 24 h prior to ICU discharge.
Table 2.
Characteristics of new onset atrial fibrillation (AF) and pharmacological management of new onset AF in intensive care unit (ICU) patients.
| New onset AF cohort | n = 3911 |
|---|---|
| Patient characteristics | |
| CHA2DS2-VASC score (median [IQR]) | 2 [1.0–3.0] |
| AF characteristics | |
| Time from admission to onset of AF (hours, median [IQR]) | 33 [12–63] |
| Number of episodes of AF during admission (median [IQR]) | 2 [1–3] |
| Duration of episodes of AF (hours, median [IQR]) | 2 [1–8] |
| % Of admission spent in AF (median [IQR]) | 9.1 [2.6–28.6] |
| Initial heart rate at onset of AF (median [IQR]) | 101 [82.0–125.0] |
| Rapid ventricular response at onset of AF (n, %) | 1994 (51.0 %) |
| Heart rate while in AF (median [IQR]) | 96 [82–112] |
| Maximum heart rate while in AF (median, [IQR]) | 122 [97–142] |
| New vasoactive requirement with first onset of AF (%) | 169 (4.3 %) |
| Sinus rhythm at ICU discharge (%), survivors, n = 3389 | 2325 (68.6 %) |
| AF at ICU discharge (%), survivors, n = 3389 | 895 (26.4 %) |
| Other recorded rhythm at ICU discharge (%), survivors, n = 3389 | 169 (5 %) |
| Absence of AF for 24 h prior to ICU discharge (%), survivors, n = 3389 | 1804 (53.2 %) |
| New onset AF pharmacological management | |
| Any rhythm control post AF onset (% yes) | 1998 (51.1 %) |
| Amiodarone administered (%) | 1971 (50.4 %) |
| Sotalol administered (%) | 51 (1.3 %) |
| Flecainide administered (%) | 0 (0 %) |
| Survivors discharged from ICU on amiodarone (%), n = 3389 | 849 (25.1 %) |
| Agents potentially used for rate control | |
| Digoxin administered post AF onset (%) | 753 (19.3 %) |
| Survivors discharged from ICU on digoxin (%), n = 3389 | 321 (9.5 %) |
| Exposure to (any) beta-blocker prior to AF onset (%) | 769 (19.7 %) |
| Exposure to (any) beta-blocker post AF onset (%) | 1425 (36.4 %) |
| Survivors discharged from ICU on a beta-blocker (%), n = 3389 | 962 (28.4 %) |
| Calcium cchannel blockers | |
| Exposure to diltiazem prior to AF onset (%) | 15 (0.4 %) |
| Exposure to diltiazem post AF onset (%) | 34 (0.9 %) |
| Survivors discharged from ICU on diltiazem (%), n = 3389 | 24 (0.7 %) |
| Exposure to verapamil prior to AF onset (%) | 14 (0.4 %) |
| Exposure to verapamil post AF onset (%) | 30 (0.8 %) |
| Survivors discharged on verapamil from ICU on diltiazem (%), n = 3389 | 15 (0.4 %) |
| Magnesium administered on day of AF onset (%) | 2634 (67.3 %) |
IQR: interquartile range.
Over half (51.1 %) of patients were prescribed a pharmacological agent for rhythm control post onset of AF, with amiodarone being prescribed to 50.4 % of patients with NOAF. Amiodarone was continued on ICU discharge for 25.1 % of all surviving patients with NOAF. Magnesium was also frequently administered on the day of onset of NOAF (67.3 %). Digoxin was prescribed to 19.3 % of patients with NOAF and 36.4 % of patients had beta-blockers prescribed during their ICU admission (see Table 2 and Table S9). The median CHA2DS2-VASC score was 2 (IQR: 1–3) (see Table 2).
Table 3 outlines outcomes for patients with and without NOAF in the ICU. Patients with NOAF had higher ICU (5 vs 2 days, p < 0.001) and hospital length of stay (12 vs 9 days, p < 0.001) than patients who did not have NOAF, as well as higher ICU (13.3 % vs 6.5 %, p < 0.001), hospital (17.2 % vs 8.5 %, p < 0.001), and 1-year mortality (26.5 vs 16.0 %, p < 0.001). After adjustment for confounders in a Cox proportional hazards model, NOAF remained a significant hazard for mortality (hazard ratio: 1.26, CI: 1.17–1.35, p < 0.001; Fig. 2, Table S4, Figure S1 and Table S7 and S8).
Table 3.
Outcomes for patients with and without new onset atrial fibrillation (AF) in the intensive care unit (ICU).
| Without AF (n = 42,688) | New Onset AF (n = 3911) | p | |
|---|---|---|---|
| Supportive therapy required | |||
| Any vasoactives during ICU (%) | 17,797 (41.7 %) | 3129 (80.0 %) | <0.001 |
| Any invasive medical ventilation (%) | 21,303 (49.9 %) | 2972 (76.0 %) | <0.001 |
| Any renal replacement therapy (%) | 1867 (4.4 %) | 766 (19.6 %) | <0.001 |
| LOS and mortality outcomes | |||
| ICU LOS (days [IQR]) | 2 [2–4] | 5 [3–10] | <0.001 |
| Hospital LOS (days [IQR]) | 9 [5–17] | 12 [7–22] | <0.001 |
| ICU mortality (%) | 2772 (6.5 %) | 522 (13.3 %) | <0.001 |
| Hospital mortality (%) | 3629 (8.5 %) | 674 (17.2 %) | <0.001 |
| 1-year mortality (%) | 6826 (16.0 %) | 1037 (26.5 %) | <0.001 |
IQR: interquartile range; LOS: length of stay. Bold text indicates a P value of <0.05.
Fig. 2.
Adjusted and unadjusted survival curves with 95 % confidence intervals for patients with and without new onset atrial fibrillation (AF) in the intensive care unit (ICU).
Differences in NOAF in patients both with and without sepsis were additionally investigated (see Table 4). Compared to patients without sepsis who developed NOAF, patients with sepsis had higher markers of illness severity. However, of those patients with NOAF who survived their ICU stay, patients with sepsis were more likely to be discharged in sinus rhythm than those without sepsis (72.2 % vs 65.7 %, p < 0.001). When adjusted for significant contributors (including illness severity, age, sex, and comorbid heart disease), the development of NOAF remains associated with both sepsis (OR: 1.67, CI: 1.52–1.85, p < 0.001) and septic shock (OR: 1.95, CI: 1.74–2.17, p < 0.001; see Table S5 and S6, Figures S1 and S2). Patients with NOAF with sepsis were more likely to receive pharmacological management with amiodarone (56.5 vs 45.3 %, p < 0.001) and digoxin (21.8 vs 17.1 %, p < 0.001) than patients with NOAF without sepsis (See Table 5). There was no difference in exposure to beta-blockers post NOAF onset in patients with and without sepsis (37.4 vs 35.6 %, p = 0.243).
Table 4.
Characteristics of new onset atrial fibrillation (AF) patients with and without sepsis during their Intensive Care Unit (ICU) Admission.
| New onset AF patients (n = 3911) | |||
|---|---|---|---|
| No sepsis (n = 2140) | Sepsis (n = 1771) | p | |
| Sex (% female) | 786 (36.7 %) | 715 (40.4 %) | 0.021 |
| Age (median [IQR]) | 64 [51.0–74.0] | 65 [52.0–74.0] | 0.154 |
| Over 60 years old (%) | 1262 (59.0 %) | 1075 (60.7 %) | 0.28 |
| Charlson Comorbidity index (median [IQR]) | 3.0 [2.0–5.0] | 4.0 [2.0–6.0] | <0.001 |
| History of ischaemic heart disease | 220 (10.3 %) | 166 (9.4 %) | 0.36 |
| History of congestive heart failure | 307 (14.3 %) | 353 (19.9 %) | <0.001 |
| Markers of illness severity | |||
| APACHE III score | 56 [40.0–77.0] | 68 [50.0–88.0] | <0.001 |
| ANZROD (% ROD [IQR]) | 4.0 [1.0–23.2] | 9.9 [2.5–29.5] | <0.001 |
| Admission SOFA score [IQR] | 6 [4.0–8.0] | 7 [5.0–10.0] | <0.001 |
| Admission SOFA score cardiovascular component [IQR] | 3 [0.0–3.0] | 3 [0.0–4.0] | <0.001 |
| Septic shock (%) | – | 913.0 (51.6 %) | |
| Day 1 highest lactate (mmol/L [IQR]) | 2.2 [1.4–4.0] | 2.6 [1.6–4.9] | <0.001 |
| Supportive therapies and LOS and mortality outcomes | |||
| Vasoactives during ICU (%) | 1570 (73.4 %) | 1559 (88.0 %) | <0.001 |
| Hydrocortisone administered after sepsis 3.0 criteria met (%) | – | 674 (38.1 %) | – |
| Invasive mechanical ventilation (%) | 1590 (74.3 %) | 1382 (78.0 %) | 0.007 |
| Renal replacement therapy (%) | 268 (12.5 %) | 498 (28.1 %) | <0.001 |
| ICU LOS (days [IQR]) | 3 [2.0–6.0] | 4 [3.0–8.5] | <0.001 |
| ICU mortality (%) | 276 (12.9 %) | 246 (13.9 %) | 0.37 |
| Hospital LOS (days [IQR]) | 10.5 [6.0–19.0] | 12.0 [7.0–23.0] | <0.001 |
| Hospital mortality (%) | 331 (15.5 %) | 343 (19.4 %) | 0.001 |
| 1-year mortality | 514 (24.0 %) | 523 (29.5 %) | <0.001 |
| AF characteristics | |||
| Day of AF onset - day sepsis 3.0 criteria met (median [IQR]) | – | +1.0 [1.0–3.0] | – |
| CHA2DS2-VASC score (median [IQR]) | 2.0 [1.0–3.0] | 2.0 [1.0–3.0] | 0.023 |
| Time from admission to onset of AF (hours, median [IQR]) | 29.0 [11.0–55.0] | 37.0 [15.0–71.0] | <0.001 |
| Number of episodes of AF during admission (median [IQR]) | 1.0 [1.0–3.0] | 2.0 [1.0–4.0] | <0.001 |
| % Of admission spent in AF (median [IQR]) | 10.0 [3.1–30.1] | 8.2 [2.2–25.7] | <0.001 |
| Initial heart rate at first onset of AF (median [IQR]) | 97.5 [79.0–120.0] | 107.0 [85.0–129.0] | <0.001 |
| Rapid ventricular rate at first AF onset (n, %) | 983.0 (45.9 %) | 1011.0 (57.1 %) | <0.001 |
| Maximum heart rate in AF (median [IQR]) | 116.0 [93.0–139.0] | 127.0 [104.0–145.5] | <0.001 |
| New vasoactive requirement with first onset of AF (%) | 90 (4.2 %) | 79 (4.5 %) | 0.694 |
| Sinus rhythm at ICU discharge, survivors, n = 3389 (%) | 1224 (65.7 %) | 1101 (72.2 %) | <0.001 |
| Sinus rhythm for 24 h prior to ICU discharge (%), survivors, n = 3389 | 746 (40 %) | 839 (55 %) | <0.001 |
ANZROD: Australian and New Zealand Risk of Death model; APACHE III: Acute Physiology and Chronic Health Evaluation III; IQR: interquartile range; LOS: length of stay; ROD: risk of death; SOFA: Sequential Organ Failure Assessment score. Bold text indicates a P value of <0.05.
Table 5.
Pharmacological management of new onset atrial fibrillation (AF) patients with and without sepsis during their intensive care unit (ICU) admission.
| Pharmacological management of new onset AF | |||
|---|---|---|---|
| No Sepsis (n = 2140) | Sepsis (n = 1771) | p | |
| Any rhythm control post AF (%) | 991 (46.3 %) | 1007 (56.9 %) | <0.001 |
| Amiodarone administered (%) | 970 (45.3 %) | 1001 (56.5 %) | <0.001 |
| Sotalol administered (%) | 32 (1.5 %) | 19 (1.1 %) | 0.261 |
| Flecainide administered (%) | 0 (0.0 %) | 0 (0.0 %) | 1 |
| Survivors discharged on amiodarone (%), n = 3389 | 468 (25.1 %) | 381 (25.0 %) | 0.937 |
| Agents potentially used for rate control | |||
| Digoxin administered post AF (%) | 367 (17.1 %) | 386 (21.8 %) | <0.001 |
| Survivors discharged on digoxin (%), n = 3389 | 176 (9.4 %) | 145 (9.5 %) | 0.953 |
| Exposure to (any) beta-blocker prior to AF onset (%) | 452 (21.1 %) | 317 (17.9 %) | 0.012 |
| Exposure to (any) beta-blocker post AF onset (%) | 762 (35.6 %) | 663 (37.4 %) | 0.243 |
| Survivors discharged on (any) beta-blocker (%), n = 3389 | 537 (28.8 %) | 425 (27.9 %) | 0.566 |
| Calcium channel blockers | |||
| Exposure to diltiazem prior to AF onset (%) | 9 (0.4 %) | 6 (0.3 %) | 0.798 |
| Exposure to diltiazem post AF onset (%) | 16 (0.7 %) | 18 (1.0 %) | 0.391 |
| Survivors discharged from ICU on diltiazem (%), n = 3389 | 11 (0.6 %) | 13 (0.9 %) | 0.413 |
| Exposure to verapamil prior to AF onset (%) | 10 (0.5 %) | 4 (0.2 %) | 0.284 |
| Exposure to verapamil post AF onset (%) | 17 (0.8 %) | 13 (0.7 %) | 0.856 |
| Survivors discharged on verapamil from ICU on diltiazem (%), n = 3389 | 10 (0.5 %) | 5 (0.3 %) | 0.442 |
| Magnesium administered on day of AF onset (%) | 1401 (65.5 %) | 1233 (69.6 %) | 0.006 |
IQR: interquartile range; LOS: length of stay. Bold text indicates a P value of <0.05.
4. Discussion
In this multicentre study of critically ill patients, NOAF was common, affecting approximately 1 in 12 ICU admissions. NOAF was associated with higher illness severity, increased prevalence of both sepsis and septic shock, and worse clinical outcomes, including prolonged length of stay and increased mortality at multiple time points. Secondly, the majority of patients with NOAF who survived their ICU admission were discharged in sinus rhythm, with those who had sepsis-associated NOAF demonstrating higher rates of reversion to sinus rhythm compared to patients without sepsis. Thirdly, after controlling for important confounders including illness severity, age, sex, and comorbid conditions, NOAF remained independently associated with increased mortality. Finally, there was a clear preference for rhythm control strategies in clinical practice, with amiodarone being the most frequently prescribed pharmacological agent, though rate control agents and other rhythm control medications were also utilised, reflecting the varied approaches to managing this arrhythmia in the critically ill population.
4.1. Incidence of NOAF in critically ill patients
NOAF has been previously shown to occur in 5 to 15 % of patients admitted to the ICU.2 Our finding that NOAF affected 8.4 % of critically ill patients falls within this reported range and provides important contemporary data from the Australian context where limited information previously existed. This consistency across international settings suggests that NOAF represents a common complication of critical illness regardless of the geographic location or healthcare system.
4.2. Association with mortality and adverse outcomes
Previous studies have consistently demonstrated that patients who develop NOAF during their ICU admission are at higher risk of both in- and post-hospital death, and have an increased risk of readmission to hospital with AF, heart failure, and thromboembolism than those who do not develop NOAF.[4], [5], [6] Our findings align with this literature, demonstrating higher ICU, hospital, and 1-year mortality in patients with NOAF. Furthermore, the association between NOAF and mortality persisted after adjustment for illness severity and other confounders. However, whether NOAF and its sequalae are directly responsible for the increased risk of mortality remains unknown.
4.3. Resolution of NOAF prior to ICU discharge
Our observation that the majority of patients with NOAF who survived their ICU stay reverted to sinus rhythm prior to discharge is consistent with the previous literature reporting that most medical and non-cardiac surgical adult ICU patients with ICU-acquired AF leave the ICU in sinus rhythm.9,12 This finding has important implications for post-ICU management and raises questions about ongoing follow-up and secondary prevention following resolution of the acute illness trigger.
4.4. Pharmacological management strategies for NOAF
We found a clear preference for amiodarone, which was prescribed to approximately half of the patients with NOAF in the ICU. Previous surveys of practice amongst ICU clinicians have had varied results; with others similarly demonstrating a preference for amiodarone10,12,23,24 and others indicating a preference for agents specifically targeting rate control (e.g. beta-blockers).25,26 The variability of practice likely reflects the limited and conflicting evidence from prospective trials27,28 and the resultant uncertainty regarding the optimal approach to managing NOAF in critically ill patients. Haemodynamic instability associated with NOAF may influence the clinician’s management strategy, and guidelines suggest the use of electrical cardioversion in this instance.29 Of note, the need for new vasoactives at the time of onset of NOAF in the present study was low (4.3 %). Further studies are required to assess the impact of NOAF, and its various management strategies, on the haemodynamics of patients already on vasoactives at the time of NOAF onset.
4.5. Sepsis-associated NOAF
Our finding that sepsis was commonly associated with NOAF aligns with established evidence that sepsis creates a pro-arrhythmogenic state.14,[30], [31], [32] Consistent with prior literature,[33], [34], [35] we observed that patients with sepsis-associated NOAF had higher illness severity and worse outcomes, including longer hospitalisation and increased mortality. Our novel finding that septic patients were more likely to revert to sinus rhythm at ICU discharge supports the hypothesis that resolution of the underlying inflammatory trigger may facilitate restoration of normal rhythm.
4.6. Study strengths and limitations
A significant strength of this study is its large, diverse critical care population, wherein granular data have been collected from multiple tertiary and non-tertiary centres. Hourly rhythm data were collected from clinical staff unbiased to the study question as part of routine clinical practice. There are also several limitations. The accuracy of AF detection was unable to be ascertained in this retrospective study. Brief episodes of AF (<1 h), or AF not associated with tachycardia (<100bpm), may have been more likely to have been missed and thus the true incidence of NOAF may be higher than described herein. Future studies investigating both the accuracy of clinician-detected AF in the ICU and the clinical significance of brief self-resolving episodes of NOAF are warranted to inform interventional trials. We were also unable to determine precise temporal relationships between medications potentially used to treat NOAF and reversion to sinus rhythm and/or normalisation of heart rate. Additional studies are thus required to assess the thresholds for treatment of NOAF in the ICU setting and the effectiveness of various treatments. Data regarding electrical cardioversion were not available in the dataset. Finally, we were unable to determine whether patients had recurrent episodes of AF in hospital subsequent to ICU discharge and what secondary or tertiary prevention strategies for AF and its sequalae were administered after ICU discharge. Future studies investigating outcomes for critical illness-associated AF should aim to include information on post-ICU outcomes.
5. Conclusions
In a large Australian multicentre study of critically ill patients, NOAF was common. The majority of patients with NOAF who survived their ICU stay reverted to sinus rhythm prior to discharge. Sepsis was more common in patients with NOAF compared to those without AF. However, patients with sepsis and NOAF were more likely to be in a sinus rhythm at ICU discharge than patients with NOAF without sepsis. Future studies are required to determine the optimal short- and long-term management strategies for AF triggered by critical illness to prevent both short- and longer-term sequalae.
Statement of ethics
This study was approved by the Metro South Hospital and Health Service Human Research Ethics Committee (HREC/2022/QMS/82024), and an individual waiver of consent was granted.
Author contributions
KD, SW, and KW contributed to conception and design.
KD, SW, JM, and KW contributed to the manuscript’s first draft.
All authors contributed to manuscript review and editing.
Data availability statement
Data cannot be shared publicly due to institutional ethics, privacy, and confidentiality regulations. Data released for research under Sect. 280 of the Public Health Act 2005 requires an application to the Director-General of Queensland Health (PHA@health.qld.gov.au).
Funding sources
This research received no specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Conflict of interest statement
None.
Acknowledgments
The authors acknowledge the Statistical Analysis and Linkage Unit of the Statistical Services Branch (SSB), Queensland Health, for linking the data sets used in this project.
The authors thank the ANZICS CORE management committee and the clinicians, data collectors, and researchers at the following contributing sites: Caboolture Hospital, Cairns Hospital, Gold Coast University Hospital, Logan Hospital, Mackay Base Hospital, Princess Alexandra Hospital, Redcliffe Hospital, Rockhampton Hospital, Royal Brisbane and Women’s Hospital, Sunshine Coast University Hospital, The Prince Charles Hospital, and the Townsville University Hospital.
Footnotes
Supplementary data to this article can be found online at https://doi.org/10.1016/j.ccrj.2026.100162.
Contributor Information
Kerina J. Denny, Email: kerina.denny@health.qld.gov.au.
the Queensland Critical Care Research Network (QCCRN):
James McCullough, Kerina J. Denny, Mandy Tallott, Andrea Marshall, Sunil Sane, Aashish Kumar, Lynette Morrison, Pam Dipplesman, Ahmad Nasser, David Stewart, Vikram Shah, Kyle White, Adam Suliman, Lachlan Quick, Jason Meyer, Ra'eesa Doola, Rod Hurford, Meg Harward, James Walsham, Karthik Venkatesh, Adam Visser, Judy Smith, Neeraj Bhadange, Vijo Kuruvilla, Kevin B. Laupland, Felicity Edwards, Jayesh Dhanani, Pierre Clement, Nermin Karamujic, Kiran Shekar, Dinesh Parmar, George Cornmell, Jayshree Lavana, Denzil Gill, Alexis Tabah, Stuart Baker, Hamish Pollock, Kylie Jacobs, Mahesh Ramanan, Prashanti Marella, Jatinder Grewal, Patrick Young, Julia Affleck, Emma Williams, Peter Garrett, Paula Lister, Vikram Masurkar, Lauren Murray, Jane Brailsford, Janine Garrett, Langa Lutshaba, Raju Pusapati, Cameron Anderson, Antony G. Attokaran, Jaco Poggenpoel, Josephine Reoch, Stephen Luke, Anni Paasilahti, Jennifer Taylor, Eamon Raith, Siva Senthuran, Stephen Whebell, Sananta Dash, Philippa McIlroy, Sebastiaan Blank, Ben Nash, Michelle Gatton, Zephanie Tyack, and Sam Keogh
Corporate Authorship
James McCullough, Kerina J Denny, Mandy Tallott, Andrea Marshall, Sunil Sane, Aashish Kumar, Lynette Morrison, Pam Dipplesman, Ahmad Nasser, David Stewart, Vikram Shah, Kyle White, Adam Suliman, Lachlan Quick, Jason Meyer, Ra'eesa Doola, Rod Hurford, Meg Harward, James Walsham, Karthik Venkatesh, Adam Visser, Judy Smith, Neeraj Bhadange, Vijo Kuruvilla, Kevin B. Laupland, Felicity Edwards, Jayesh Dhanani, Pierre Clement, Nermin Karamujic, Kiran Shekar, Dinesh Parmar, George Cornmell, Jayshree Lavana, Denzil Gill, Alexis Tabah, Stuart Baker, Hamish Pollock, Kylie Jacobs, Mahesh Ramanan, Prashanti Marella, Jatinder Grewal, Patrick Young, Julia Affleck, Emma Williams, Peter Garrett, Paula Lister, Vikram Masurkar, Lauren Murray, Jane Brailsford, Janine Garrett, Langa Lutshaba, Raju Pusapati, Cameron Anderson, Antony G. Attokaran, Jaco Poggenpoel, Josephine Reoch, Stephen Luke, Anni Paasilahti, Jennifer Taylor, Eamon Raith, Siva Senthuran, Stephen Whebell, Sananta Dash, Philippa McIlroy, Sebastiaan Blank, Ben Nash, Michelle Gatton, Zephanie Tyack, and Sam Keogh.
Appendix A. Supplementary data
The following is the Supplementary data to this article:
References
- 1.Van Gelder I.C., Rienstra M., Bunting K.V., Casado-Arroyo R., Caso V., Crijns H., et al. 2024 ESC guidelines for the management of atrial fibrillation developed in collaboration with the European Association for Cardio-Thoracic Surgery (EACTS) Eur Heart J. 2024;45(36):3314–3414. doi: 10.1093/eurheartj/ehae176. [DOI] [PubMed] [Google Scholar]
- 2.Yoshida T., Fujii T., Uchino S., Takinami M. Epidemiology, prevention, and treatment of new-onset atrial fibrillation in critically ill: a systematic review. J Intensive Care. 2015;3(1):19. doi: 10.1186/s40560-015-0085-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Meierhenrich R., Steinhilber E., Eggermann C., Weiss M., Voglic S., Bogelein D., et al. Incidence and prognostic impact of new-onset atrial fibrillation in patients with septic shock: a prospective observational study. Crit Care. 2010;14(3) doi: 10.1186/cc9057. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Wetterslev M., Haase N., Hassager C., Belley-Cote E.P., McIntyre W.F., An Y., et al. New-onset atrial fibrillation in adult critically ill patients: a scoping review. Intensive Care Med. 2019;45(7):928–938. doi: 10.1007/s00134-019-05633-x. [DOI] [PubMed] [Google Scholar]
- 5.Garside T., Bedford J.P., Vollam S., Gerry S., Rajappan K., Watkinson P.J. Increased long-term mortality following new-onset atrial fibrillation in the intensive care unit: a systematic review and meta-analysis. J Crit Care. 2022;72 doi: 10.1016/j.jcrc.2022.154161. [DOI] [PubMed] [Google Scholar]
- 6.Bedford J.P., Ferrando-Vivas P., Redfern O., Rajappan K., Harrison D.A., Watkinson P.J., et al. New-onset atrial fibrillation in intensive care: epidemiology and outcomes. Eur Heart J Acute Cardiovasc Care. 2022;11(8):620–628. doi: 10.1093/ehjacc/zuac080. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Sibley S., Atzema C., Balik M., Bedford J., Conen D., Garside T., et al. Research priorities for the study of atrial fibrillation during acute and critical illness: recommendations from the Symposium on Atrial Fibrillation in Acute and Critical Care. BMC Proc. 2024;18(Suppl 22):23. doi: 10.1186/s12919-024-00309-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.O'Bryan L.J., Redfern O.C., Bedford J., Petrinic T., Young J.D., Watkinson P.J. Managing new-onset atrial fibrillation in critically ill patients: a systematic narrative review. BMJ Open. 2020;10(3) doi: 10.1136/bmjopen-2019-034774. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Kanji S., Williamson D.R., Yaghchi B.M., Albert M., McIntyre L. Canadian critical care trials G. Epidemiology and management of atrial fibrillation in medical and noncardiac surgical adult intensive care unit patients. J Crit Care. 2012;27(3):326 e1–8. doi: 10.1016/j.jcrc.2011.10.011. [DOI] [PubMed] [Google Scholar]
- 10.Johnston B.W., Udy A.A., McAuley D.F., Mogk M., Welters I.D., Sibley S. An international survey of the management of atrial fibrillation in critically unwell patients. Crit Care Explor. 2024;6(4) doi: 10.1097/CCE.0000000000001069. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Wetterslev M., Moller M.H., Granholm A., Hassager C., Haase N., Aslam T.N., et al. Management of acute atrial fibrillation in the intensive care unit: an international survey. Acta Anaesthesiol Scand. 2022;66(3):375–385. doi: 10.1111/aas.14007. [DOI] [PubMed] [Google Scholar]
- 12.Pardo E., Futier E., Muller L., Besch G., Vardon-Bounes F., Kipnis E., et al. Management of new-onset atrial fibrillation in critically ill patients: a national multicenter prospective cohort. Anaesth Crit Care Pain Med. 2025 doi: 10.1016/j.accpm.2025.101617. [DOI] [PubMed] [Google Scholar]
- 13.Sibley S., Bedford J., Wetterslev M., Johnston B., Garside T., Kanji S., et al. Atrial fibrillation in critical illness: state of the art. Intensive Care Med. 2025;51(5):904–916. doi: 10.1007/s00134-025-07895-0. [DOI] [PubMed] [Google Scholar]
- 14.Bedford J.P., Ede J., Watkinson P.J. Triggers for new-onset atrial fibrillation in critically ill patients. Intensive Crit Care Nurs. 2021;67 doi: 10.1016/j.iccn.2021.103114. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Maesen B., Nijs J., Maessen J., Allessie M., Schotten U. Post-operative atrial fibrillation: a maze of mechanisms. Europace. 2012;14(2):159–174. doi: 10.1093/europace/eur208. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Moss T.J., Calland J.F., Enfield K.B., Gomez-Manjarres D.C., Ruminski C., DiMarco J.P., et al. New-onset atrial fibrillation in the critically ill. Crit Care Med. 2017;45(5):790–797. doi: 10.1097/CCM.0000000000002325. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Lip G.Y., Nieuwlaat R., Pisters R., Lane D.A., Crijns H.J. Refining clinical risk stratification for predicting stroke and thromboembolism in atrial fibrillation using a novel risk factor-based approach: the euro heart survey on atrial fibrillation. Chest. 2010;137(2):263–272. doi: 10.1378/chest.09-1584. [DOI] [PubMed] [Google Scholar]
- 18.Quan H., Sundararajan V., Halfon P., Fong A., Burnand B., Luthi J.C., et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care. 2005;43(11):1130–1139. doi: 10.1097/01.mlr.0000182534.19832.83. [DOI] [PubMed] [Google Scholar]
- 19.Sundararajan V., Henderson T., Perry C., Muggivan A., Quan H., Ghali W.A. New ICD-10 version of the Charlson comorbidity index predicted in-hospital mortality. J Clin Epidemiol. 2004;57(12):1288–1294. doi: 10.1016/j.jclinepi.2004.03.012. [DOI] [PubMed] [Google Scholar]
- 20.White K.C., Serpa-Neto A., Hurford R., Clement P., Laupland K.B., See E., et al. Sepsis-associated acute kidney injury in the intensive care unit: incidence, patient characteristics, timing, trajectory, treatment, and associated outcomes. A multicenter, observational study. Intensive Care Med. 2023;49(9):1079–1089. doi: 10.1007/s00134-023-07138-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.White K.C., Costa-Pinto R., Blank S., Whebell S., Quick L., Luke S., et al. Effect of early adjunctive vasopressin initiation for septic shock patients: a target trial emulation. Crit Care. 2025;29(1):188. doi: 10.1186/s13054-025-05401-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Singer M., Deutschman C.S., Seymour C.W., Shankar-Hari M., Annane D., Bauer M., et al. The third international consensus definitions for sepsis and septic shock (Sepsis-3) JAMA. 2016;315(8):801–810. doi: 10.1001/jama.2016.0287. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Wetterslev M., Hylander Moller M., Granholm A., Hassager C., Haase N., Lange T., et al. Atrial fibrillation (AFIB) in the ICU: incidence, risk factors, and outcomes: the international AFIB-ICU cohort study. Crit Care Med. 2023;51(9):1124–1137. doi: 10.1097/CCM.0000000000005883. [DOI] [PubMed] [Google Scholar]
- 24.Chean C.S., McAuley D., Gordon A., Welters I.D. Current practice in the management of new-onset atrial fibrillation in critically ill patients: a UK-wide survey. PeerJ. 2017;5 doi: 10.7717/peerj.3716. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Moskowitz A., Chen K.P., Cooper A.Z., Chahin A., Ghassemi M.M., Celi L.A. Management of atrial fibrillation with rapid ventricular response in the intensive care unit: a secondary analysis of electronic health record data. Shock. 2017;48(4):436–440. doi: 10.1097/SHK.0000000000000869. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Walkey A.J., Evans S.R., Winter M.R., Benjamin E.J. Practice patterns and outcomes of treatments for atrial fibrillation during sepsis: a propensity-matched cohort study. Chest. 2016;149(1):74–83. doi: 10.1378/chest.15-0959. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Andreasen A.S., Wetterslev M., Sigurdsson M.I., Bove J., Kjaergaard J., Aslam T.N., et al. New-onset atrial fibrillation in critically ill adult patients-an SSAI clinical practice guideline. Acta Anaesthesiol Scand. 2023;67(8):1110–1117. doi: 10.1111/aas.14262. [DOI] [PubMed] [Google Scholar]
- 28.Drikite L., Bedford J.P., O'Bryan L., Petrinic T., Rajappan K., Doidge J., et al. Treatment strategies for new onset atrial fibrillation in patients treated on an intensive care unit: a systematic scoping review. Crit Care. 2021;25(1):257. doi: 10.1186/s13054-021-03684-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Chyou J.Y., Barkoudah E., Dukes J.W., Goldstein L.B., Joglar J.A., Lee A.M., et al. Atrial fibrillation occurring during acute hospitalization: a scientific statement from the American heart association. Circulation. 2023;147(15):e676–e698. doi: 10.1161/CIR.0000000000001133. [DOI] [PubMed] [Google Scholar]
- 30.Scott L., Jr., Li N., Dobrev D. Role of inflammatory signaling in atrial fibrillation. Int J Cardiol. 2019;287:195–200. doi: 10.1016/j.ijcard.2018.10.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Ho W.L., Umais M., Bai M., Dang N.B., Kumari K., Izhar S., et al. Beyond the beat: a multifaceted review of atrial fibrillation in sepsis: risk factors, management strategies, and economic impact. Cardiol Res. 2025;16(1):1–14. doi: 10.14740/cr1723. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Chen P.S., Chen L.S., Fishbein M.C., Lin S.F., Nattel S. Role of the autonomic nervous system in atrial fibrillation: pathophysiology and therapy. Circ Res. 2014;114(9):1500–1515. doi: 10.1161/CIRCRESAHA.114.303772. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Klein Klouwenberg P.M., Frencken J.F., Kuipers S., Ong D.S., Peelen L.M., van Vught L.A., et al. Incidence, predictors, and outcomes of new-onset atrial fibrillation in critically ill patients with sepsis. A cohort study. Am J Respir Crit Care Med. 2017;195(2):205–211. doi: 10.1164/rccm.201603-0618OC. [DOI] [PubMed] [Google Scholar]
- 34.Huo Y., Yoshimura H., Gonzalez-Izquierdo A., Lip G.Y.H., Schmidt F., Providencia R. Risk factors and prognosis of new-onset atrial fibrillation in sepsis: a nationwide electronic health record study. JACC Adv. 2025;4(4) doi: 10.1016/j.jacadv.2025.101681. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Walkey A.J., Wiener R.S., Ghobrial J.M., Curtis L.H., Benjamin E.J. Incident stroke and mortality associated with new-onset atrial fibrillation in patients hospitalized with severe sepsis. JAMA. 2011;306(20):2248–2254. doi: 10.1001/jama.2011.1615. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data cannot be shared publicly due to institutional ethics, privacy, and confidentiality regulations. Data released for research under Sect. 280 of the Public Health Act 2005 requires an application to the Director-General of Queensland Health (PHA@health.qld.gov.au).