Prognostic impact of oral anticoagulation therapy and atrial fibrillation in patients with type B acute aortic dissection

Abstract Background The prognostic impact of atrial fibrillation (AF) and oral anticoagulation (OAC) therapy in patients with type B acute aortic dissection (AAD) remains unclear. Therefore, we investigated the prognostic impact of AF and OAC therapy in patients with type B AAD. Methods Consecutive patients diagnosed with AAD were included in this single‐center, retrospective study. Patients with type B AAD were selected from the study population and divided into three groups: AF(+)/OAC(+), AF(+)/OAC(−), and AF(−)/OAC(−). The primary end point was major adverse cardiovascular and cerebrovascular events (MACCE), including all‐cause death, progressive aortic events, cerebral infarction, and organ malperfusion. Results In total, 139 patients diagnosed with type B AAD were analyzed. AF was observed in 27 patients (19%). Among them, 13 patients (9%) received OAC therapy for AF. MACCE occurred in 32 patients (23%) during the observation period: all‐cause death in four patients, progressive aortic events in 24 patients, cerebral infarction events in two patients, and malperfusion events in two patients. The incidence of MACCE was higher in the AF(+)/OAC(+) group than in the AF(+)/OAC(−) group (hazard ratio[HR]: 3.875; 95% confidence interval [CI]: 1.153–17.496). In contrast, there was no significant difference in the incidence of MACCE between the AF(+)/OAC(−) and AF(−)/OAC(−) groups (HR: 1.001, 95% CI: 0.509–1.802). Conclusion Among patients with type B AAD, the use of OAC for AF was associated with a higher risk of MACCE.


| INTRODUC TI ON
][6][7][8][9][10] However, research on whether the presence of AF contributes to a worse prognosis in patients with AAD is lacking.Furthermore, despite the importance of oral anticoagulation (OAC) therapy for preventing cardiogenic thromboembolism in patients with AF, there are no specific recommendations for OAC management in patients with AAD and AF.Although there are several reports on the safety of OAC in preventing thromboembolism in patients with AAD, 3,[11][12][13] comprehensive data on thromboembolic events and the risk of worsening dissection are limited.
Therefore, this study aims to assess the prognostic impact of AF and OAC therapy in patients with AAD.

| Study population and diagnosis criteria
We performed a retrospective analysis of consecutive patients who were diagnosed with AAD at Kitasato University Hospital between 2010 and 2019.Since the prognosis and indication for surgery for AAD differed between Stanford classification types A and B, patients with type A AAD were excluded, and only patients diagnosed with type B AAD were included in this study.Additionally, patients with missing data (follow-up data, history, and medication history), patients with AF onset after discharge, and those receiving OAC for mechanical valves were excluded (Figure 1).
5][16] The diagnosis was based on computed tomography (CT) findings, which confirmed a dissected descending aorta.Based on clinical features, all cases were classified into the nonthrombosed type, which was characterized by partial enhancement of the false lumen; and the thrombosed type, which was characterized by a false lumen without enhancement, including an intramural hematoma of the aorta. 16 was defined as an irregular ventricular rate and the absence of P waves on a standard 12-leads electrocardiogram (ECG) or lasting >30 s on a continuously monitored ECG.In this study, AF was defined as the diagnosis of AF during or before hospitalization.
Patients with new-onset AF after hospital discharge were excluded.
The types of AF were classified according to their temporal patterns: paroxysmal or nonparoxysmal AF.Paroxysmal AF was defined as spontaneous termination within 7 days. 17All other patients were defined as having nonparoxysmal AF.New-onset AF was defined as newly detected AF during hospitalization without a history of AF.
The indication of OAC therapy was at the discretion of the physician according to the guideline recommendations. 18OAC use was defined as an oral prescription during the hospital stay or prior to admission.All patients were divided into three groups based on the presence or absence of AF and the use of OAC: AF(+)/OAC(+), AF(+)/OAC(−), and AF(−)/OAC(−).

| Baseline characteristics and clinical outcomes
Baseline patient characteristics were retrospectively assessed, including age, gender, body mass index, medical history, laboratory data, imaging results, and electrocardiograms.The primary end point was defined as major adverse cardiovascular and cerebrovascular events (MACCE), which included all-cause death, progressive aortic events, organ malperfusion, and cerebral infarction during the observation period.Progressive aortic events were defined as aortic rupture, enlargement of the aortic dimension (>60 mm), recurrence of dissection, and early aortic expansion (≥1 mm/day), which were indications for surgical treatment. 14,19These events were diagnosed by CT imaging.The day of AAD onset was estimated based on the symptoms related to F I G U R E 1 Study population.In total, 307 patients diagnosed with AAD were enrolled in this study.Among them,139 patients diagnosed with type B AAD were analyzed.The patients were divided into three groups based on the presence or absence of AF and the use of OAC: AF(+)/ OAC(+), AF(+)/OAC(−), and AF(−)/OAC(−).AAD, acute aortic dissection; AF, atrial fibrillation; OAC, oral anticoagulation.AAD, or the first day on which AAD was diagnosed using CT if the symptoms were unclear.
After the initial event, patients underwent CT imaging 48 h after onset, and follow-up CT scans were performed at 6 months and 1 year after onset.Alternatively, additional CT scans were performed according to subjective symptoms and other findings of concern.The use of contrast media was not mandatory for any CT scan.
The observational period was set at 1 year after AAD onset.

| Statistical analysis
The statistical analyses were conducted using the JMP® version

| Study flowchart and comparison of patients' characteristics
In total, 307 patients diagnosed with AAD were enrolled in this study.
One hundred fifty-one patients were diagnosed with type B AAD.
Five patients were excluded because of the absence of data or loss of follow-up.Four patients with AF onset after discharge and three who underwent OAC for mechanical valves were excluded (Figure 1).

| Clinical outcomes of patients with type B AAD during the 1-year observation
The MACCE was observed in 32 patients: all-cause death in four patients, progressive aortic events in 24 patients, cerebral infarction events in two patients, and malperfusion events in two patients.Of the 24 patients with progressive aortic events, aortic rupture occurred in seven (29%), dissection recurred in three (13%), enlargement of the aortic dimension occurred in 10 (42%), and early aortic expansion occurred in four (17%) (Figure S1).
An adjusted analysis using factors obtained from the univariate analysis, such as maximum descending aortic diameter and age, was performed in addition to AF and OAC.After adjustment, the incidence of MACCE was higher in the AF(+)/OAC(+) group than in the AF(+)/ OAC(−) group (HR: 4.130, 95% CI: 1.010-21.831,p = .0483;Table 4).
In the subgroup analysis, OAC use for AF had a poor prognosis in the patients with the nonthrombosed type (HR: 7.184, 95% CI: 2.052-24.103; Figure 3).

| Prognostic impact of AF for AAD
2][3] In this study, AF was observed in 27/139 patients (19%), similar to the findings of previous studies.A study based on a Taiwanese national database reported that patients with AF had a 1.18-fold higher incidence of AAD than those without AF.between the AF(+)/OAC(−) and AF(−)/OAC(−) groups.Therefore, we concluded that AF itself could not be an independent predictor of MACCE.Interestingly, factors associated with the incidence of AF-age, hypertension, and diabetes mellitus-overlap with factors of atherosclerosis progression.It may suggest that the presence of AF is a phenotype and not an exacerbation factor of advanced atherosclerosis or AAD progression.

| Effect of OAC therapy for AF in patients with AAD
As shown in Table 1, patients with nonparoxysmal AF received OAC therapy.Although not statistically significant, there was a trend toward more preexisting hypertension in the AF (+)/OAC (+) group than in the other groups.Table S1 shows the details of the patients taking OAC.Three of the patients had no history of hypertension.The CHADS 2 score was ≥1 for all patients taking OAC.Low-risk patients were not prescribed OAC.Although the CHADS 2 score was higher in the AF(+)/OAC(+) group than in the other groups, the use of OAC was determined a risk factor for MACCE in the analysis adjusted only for the CHADS 2 score (data not shown).In this study, the incidence of MACCE was associated with the use of OAC for AF.1][12][13] The efficacy of OAC therapy in patients with type B AAD and AF remains controversial.
AF is a potent risk factor for cardiogenic thromboembolism.Additionally, the subgroup analysis suggested that the prevalence of the nonthrombosed type could be a worsening factor in patients using OAC (Figure 3).Of the 13 patients on OAC, nine experienced a MACCE event.Furthermore, five patients on OAC with MACCE events were nonthrombosed: they had a false lumen communicating with the true lumen on follow-up CT scans (Table S1).
A previous report showed that residual false lumen blood flow is a prognostic factor for AAD. 28These results suggest that the use of OAC disturbs thrombotic formation in the false lumen during the acute phase.The use of OAC may be a clinical dilemma for patients with AAD and AF who have a high risk of thrombosis.Concerning this issue, the left atrial appendage exclusion using an epicardial clip device may be suitable for avoiding OAC therapy for AAD patients with high CHA 2 DS 2 -VASc score. 29Moreover, since thoracic endovascular aortic repair was effective in patients with chronic aortic dissection in the INSTEAD XT trial, 30 it would also apply to patients with chronic nonthrombosed type aortic dissection.

| LI M ITATI O N S
This study had several limitations.First, the study population was small as this was a single-center clinical study.Moreover, the observational period was limited to 1 year; therefore, the long-term outcomes remain unclear.Second, co-occurrence with AF was defined as the presence of AF on admission or during hospitalization.We did not evaluate patients who developed AF after discharge.Therefore, the impact of AF and OAC use on aortic dissection in the chronic phase has not been assessed.Third, the indication for OAC use was based on the CHADS 2 score, but the attending physician made the final decision.In addition, patients receiving OACs may have more regular outpatient visits.These factors might have introduced a selection bias.Fourth, it was not possible to determine which type of patients with AF and aortic dissection should be prescribed OACs.

| CON CLUS IONS
The use of OAC for AF was associated with a higher risk of MACCE in patients with type B AAD and AF.These findings highlight the contentious indications for OAC therapy in patients with type B AAD F I G U R E 3 Subgroup analysis of MACCE.Hazard ratios were calculated using the Cox logistic regression analysis.OAC use was associated with a higher risk in nonthrombosed patients than in thrombosed patients.95% CI, 95% confidence interval; AF, atrial fibrillation; HR, hazard ratio; MACCE, major adverse cardiovascular and cerebrovascular events; OAC, oral anticoagulation.
of data and loss of follow-up, n=5 Patients who onset AF after discharge, n=4 Patients who use OAC for mechanical valve, n=3 Patients who were diagnosed with AAD between 2010 and 2019 in our hospital,

Finally, we could
not elucidate the detailed mechanisms of MACCE caused by OAC use.Further studies are needed to evaluate the impact of OAC therapy on MACCE in patients with type B AAD in a larger population.
Comparison of baseline characteristics between each group.
TA B L E 1 Risk factor for MACCE.
25[21][22][23][24]cardiovascular or fatal events occurred in 22% of patients, even in those with type B AAD; advanced age, female, renal dysfunction, initial aortic enlargement, and high C-reactive protein levels are poor prognostic factors in such patients.1,[21][22][23][24]There isack of data regarding whether the presence of AF contributes to a worsening prognosis in patients with type B AAD. Campia et al. reported that AF is a poor prognostic event in patients with acute aortic syndrome.25Inthisreport, 60% of patients with AF were prescribed OAC; however, the impact of OAC use was not addressed in the analysis.In this study, univariate analysis revealed that AF, OAC, age, and maximum descending aortic diameter were prognostic factors for MACCE.OAC was prescribed to 48% of the patients with AF in this cohort.Thus, it can be extrapolated that the presence of AF and the use of OAC may have an interaction effect.Additionally, we could not perform multivariate analysis, because of the small sample size.This study found no significant difference in MACCE TA B L E 2 *Asterisk indicates significance.TA B L E 3 Primary outcomes among each group. Abbrevition: MACCE, major adverse major adverse cardiovascular and cerebrovascular events.*Asterisk indicates significance.
Primary outcomes adjusted by each parameter.the factors constituting the CHA 2 DS 2 -VASc score, and patients with AAD often have hypertension; therefore, the CHA 2 DS 2 -VASc score of patients with AAD and AF would be >2, resulting in these *Asterisk indicates significance. of