Article Text

Download PDFPDF

Original research
Early ischaemic and haemorrhagic complications after atrial fibrillation-related ischaemic stroke: analysis of the IAC study
  1. Shadi Yaghi1,
  2. Nils Henninger2,
  3. Erica Scher1,
  4. James Giles3,
  5. Angela Liu3,
  6. Muhammad Nagy2,
  7. Ashutosh Kaushal4,
  8. Idrees Azher4,5,
  9. Brian Mac Grory4,
  10. Hiba Fakhri6,
  11. Kiersten Brown Espaillat6,
  12. Syed Daniyal Asad7,
  13. Hemanth Pasupuleti8,
  14. Heather Martin8,
  15. Jose Tan8,
  16. Manivannan Veerasamy8,
  17. Ava L Liberman5,
  18. Charles Esenwa5,
  19. Natalie Cheng5,
  20. Khadean Moncrieffe5,
  21. Iman Moeini-Naghani9,
  22. Mithilesh Siddu9,
  23. Tushar Trivedi1,
  24. Christopher R Leon Guerrero9,
  25. Muhib Khan8,10,
  26. Amre Nouh7,
  27. Eva Mistry6,
  28. Salah Keyrouz3,
  29. Karen Furie4
  1. 1 Department of Neurology, NYU Langone Health, New York, New York, USA
  2. 2 Department of Neurology, University of Massachusetts Medical School, Worcester, Massachusetts, USA
  3. 3 Department of Neurology, Washington University in Saint Louis, St Louis, Missouri, USA
  4. 4 Department of Neurology, Brown University Warren Alpert Medical School, Providence, Rhode Island, USA
  5. 5 Department of Neurology, Montefiore Hospital and Medical Center, Bronx, New York, USA
  6. 6 Department of Neurology, Vanderbilt University Medical Center, Nashville, Tennessee, USA
  7. 7 Department of Neurology, Hartford Hospital, Hartford, Connecticut, USA
  8. 8 Department of Neurology, Spectrum Health, Grand Rapids, Michigan, USA
  9. 9 Department of Neurology, The George Washington University School of Medicine and Health Sciences, Washington, District of Columbia, USA
  10. 10 Department of Neurology, Michigan State University College of Human Medicine, East Lansing, Michigan, USA
  1. Correspondence to Dr Shadi Yaghi, Neurology, NYU Langone Health, New York, NY 10016, USA; shadiyaghi{at}yahoo.com

Abstract

Introduction Predictors of long-term ischaemic and haemorrhagic complications in atrial fibrillation (AF) have been studied, but there are limited data on predictors of early ischaemic and haemorrhagic complications after AF-associated ischaemic stroke. We sought to determine these predictors.

Methods The Initiation of Anticoagulation after Cardioembolic stroke study is a multicentre retrospective study across that pooled data from consecutive patients with ischaemic stroke in the setting of AF from stroke registries across eight comprehensive stroke centres in the USA. The coprimary outcomes were recurrent ischaemic event (stroke/TIA/systemic arterial embolism) and delayed symptomatic intracranial haemorrhage (d-sICH) within 90 days. We performed univariate analyses and Cox regression analyses including important predictors on univariate analyses to determine independent predictors of early ischaemic events (stroke/TIA/systemic embolism) and d-sICH.

Results Out of 2084 patients, 1520 patients qualified; 104 patients (6.8%) had recurrent ischaemic events and 23 patients (1.5%) had d-sICH within 90 days from the index event. In Cox regression models, factors associated with a trend for recurrent ischaemic events were prior stroke or transient ischemic attack (TIA) (HR 1.42, 95% CI 0.96 to 2.10) and ipsilateral arterial stenosis with 50%–99% narrowing (HR 1.54, 95% CI 0.98 to 2.43). Those associated with sICH were male sex (HR 2.68, 95% CI 1.06 to 6.83), history of hyperlipidaemia (HR 2.91, 95% CI 1.08 to 7.84) and early haemorrhagic transformation (HR 5.35, 95% CI 2.22 to 12.92).

Conclusion In patients with ischaemic stroke and AF, predictors of d-sICH are different than those of recurrent ischaemic events; therefore, recognising these predictors may help inform early stroke versus d-sICH prevention strategies.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

Cardioembolic stroke in patients with atrial fibrillation (AF) is associated with high mortality and morbidity,1 carrying a relatively high risk of early ischaemic recurrence and haemorrhagic transformation.2 3 Initiating anticoagulation acutely after a cardioembolic stroke is typically avoided due to the risk of early haemorrhagic complications and current guidelines suggest starting anticoagulation in the 4–14 days interval after the index stroke.4

Predictors of long-term ischaemic and haemorrhagic complications in patients with AF have been extensively studied,5 but there are very limited data on predictors of early ischaemic and haemorrhagic complications after an AF-associated cardioembolic stroke.6 Determining these predictors may help clinicians individualise treatments based on differential ischaemic and haemorrhagic risks, and may help determine the optimal timing of initiation of anticoagulation based on their early haemorrhagic and ischaemic risk.

In this study, we sought to determine clinical, laboratory and imaging predictors of early ischaemic and non-acute therapy-related haemorrhagic events in patients with ischaemic stroke in the setting of AF.

Methods

Study sample

The Initiation of Anticoagulation after Cardioembolic stroke (IAC) study is a multicentre retrospective study across that pooled data from stroke registries across eight comprehensive stroke centres in the USA. Consecutive patients hospitalised with acute ischaemic stroke in the setting of AF (history of AF or AF diagnosed on admission) were included. Patients who developed AF after the index hospitalisation were excluded. In addition, patients with competing stroke mechanisms were included in our study consistent with the main goal of the IAC study which sought to determine the optimal timing of initiating anticoagulation and because these patients would typically be treated with anticoagulation for secondary prevention regardless of the primary stroke mechanism.

Study variables

The following study variables were collected.

Demographic factors

Age and sex.

Clinical variables

Time from onset to hospital arrival, comorbidities (hypertension, diabetes, hyperlipidaemia, prior stroke or TIA, congestive heart failure, coronary artery disease, peripheral vascular disease, smoking, CHA2DS2-Vasc score), NIHSS score and systolic and diastolic blood pressures on admission.

Medications prior to admission

Antiplatelet use, anticoagulant use and statin use.

Laboratory values on admission

Glucose level, platelet count and low-density lipoprotein (LDL).

Neuroimaging and vascular imaging variables

Presence of intracranial or extracranial atherosclerosis with ≥50% luminal narrowing in the territory of the stroke, location of stroke (anterior circulation vs posterior circulation), largest ischaemic stroke lesion volume (moderate to large defined as >10 mL) and early haemorrhagic transformation in the setting of acute therapy occurring within 48 hours from index event.7 8

Echocardiographic variables

Left atrial enlargement (determined by left atrial diameter, volume, index or volume index),9 moderate to severe aortic or mitral valve heart disease, bioprosthetic valve, intracranial thrombus, spontaneous echocardiographic contrast and ejection fraction.

In-hospital treatments

Alteplase and mechanical thrombectomy, anticoagulation initiation and time to initiate anticoagulation.

Outcomes

The study coprimary outcomes were recurrent ischaemic event (stroke/TIA/systemic arterial embolism) and delayed symptomatic intracranial haemorrhage (d-sICH) within 90 days. D-sICH was defined as neurological deterioration in the setting of any new or worsening haemorrhage and the haemorrhage being the cause of the neurological deterioration.7 Early haemorrhagic transformation related to acute stroke therapy was not considered as outcome but these patients were still included in the analysis.

In all participating centres, patients discharged with a diagnosis of stroke were scheduled to have an in-person clinic visit at 90 days. In addition, in three out of eight centres, prespecified phone calls were performed at the 30 days (in one centre) and 90 days (in two centres) time points that assessed for recurrent ischaemic and haemorrhagic outcomes. Outcomes were preferentially abstracted from the 90-day patient follow-up visit and prespecified 90-day phone calls. In the sites assessing 90-day outcomes by phone and for patients not showing up to their 90-day visits, three attempts were made on different occasions to contact the patient or healthcare provider by phone. If unsuccessful, then outcomes were assessed by chart review of hospitalisation and other outpatient visit and outside hospital records. All outcomes were abstracted by the study local research assistant and confirmed by the site principal investigator. Multiple queries were sent to the participating sites regarding study outcomes and other variables in our dataset and several data cross-checks were performed to confirm the integrity of the data sent by individual sites.

Analytical plan

Data from sites were pooled and queries were sent to assure accuracy of data, as indicated. We excluded patients who were lost to follow-up as well as those who did not experience the study outcome of interest but died within 90 days. We stratified patients based on the occurrence of the coprimary outcomes (recurrent ischaemic events or d-sICH) and compared the rates of the respective primary outcomes across the above-mentioned study variables. We then performed Cox regression analyses including important predictors found on univariate analyses to determine independent predictors of early ischaemic events and d-sICH, respectively. Analyses were done using SPSS V.25.0.

Results

We included 2084 patients into the parent study. Of these, 564 patients were excluded due to: 195 lost to follow-up and 369 died within 90 days without having a study outcome. The final sample included 1520 patients; median age was 78 years (68–86) and 50.1% (762/1520) were women. Among the included patients, 104 patients (6.8%) had recurrent ischaemic events (91 patients with recurrent stroke, 6 patients with TIA and 9 patients with systemic embolism) and 23 patients (1.5%) had a d-sICH within 90 days from the index event (figure 1).

Univariate analyses

In univariate analyses, patients with recurrent ischaemic events were more likely to have a history of prior stroke or TIA (39.4% vs 30.7%, p=0.079) and ipsilateral arterial stenosis with 50%–99% narrowing (23.1% vs 15.9%, p=0.073) as compared with patients without recurrent events. Other clinical, laboratory and radiographic variables did not differ significantly between groups (table 1).

Table 1

Univariate analyses of predictors of recurrent ischaemic events

When compared with subjects without d-sICH, patients with d-sICH were more likely to be men (73.9% vs 49.5%, p=0.021), have hyperlipidaemia (78.3% vs 55.8%, p=0.034), a largest ischaemic lesion size >10 mL (82.6% vs 58.5%, p=0.019) and early haemorrhagic transformation (60.9% vs 17.0%, p<0.001) (table 2).

Table 2

Univariate analyses of predictors of delayed symptomatic intracranial haemorrhage (d-sICH)

Multivariable models for variables associated with recurrent ischemic events

In Cox regression models, factors with a trend for increased odds of recurrent ischaemic events were prior stroke and TIA (adjusted HR 1.42, 95% CI 0.96 to 2.10, p=0.082) and ipsilateral arterial stenosis with 50%–99% narrowing (adjusted OR 1.54, 95% CI 0.98 to 2.43, p=0.063) (table 3). Survival analyses of these predictors are shown in figure 2.

Figure 2

Kaplan-Meier survival analysis showing predictors of recurrent ischaemic events stratified by (A). Prior stroke or TIA and (B). Ipsilateral atherosclerosis with 50%–99% luminal narrowing.

Table 3

Cox regression model showing predictors of ischaemic events

Multivariable models for variables associated with sICH

In Cox regression models, factors associated with d-sICH were male sex (adjusted HR 2.68, 95% CI 1.06 to 6.83, p=0.038), history of hyperlipidaemia (adjusted HR 2.91, 95% CI 1.08 to 7.84, p=0.035) and early haemorrhagic transformation (HR 5.35, 95% CI 2.22 to 12.92, p<0.001) (table 4). Sensitivity analyses adding age to the Cox regression model showed no significant associations between age and d-sICH (adjusted HR 0.98, 95% CI 0.95 to 1.02, p=0.338) and the rest of the associations remained unchanged (table 4). Survival analyses of these predictors are shown in figure 3.

Figure 3

Kaplan-Meier survival analysis showing predictors of delayed symptomatic intracranial haemorrhage (sICH) stratified by (A). Sex, (B). History of hyperlipidaemia and (C). Presence of early haemorrhagic transformation.

Table 4

Cox regression models of predictors of symptomatic intracranial haemorrhage

Additional analysis

We performed additional analyses to better understand the association between hyperlipidaemia and d-sICH. In these analyses, when compared with patients without a history of hyperlipidaemia, those with hyperlipidaemia were more likely to be on statins (68.7% vs 37.4%, p<0.001). In addition, the median (IQR) LDL was lower in those with a history of hyperlipidaemia vs those without (75 (42) vs 80 (43), p=0.004). This suggests increased usage of statins and lower LDL cholesterols in those with versus without history of hyperlipidaemia.

Moreover, given that ipsilateral atherosclerosis may be more associated with recurrent ischaemic stroke or TIA, we performed additional analysis looking at the association between these two. In these analyses, there was significant association between ipsilateral atherosclerosis causing 50% or more arterial narrowing and recurrent ischaemic stroke or TIA (9.4% vs 5.8%, p=0.044).

Discussion

In this large multicentre study, we identified predictors of early recurrent ischaemic events and d-sICH in patients with cardioembolic stroke in the setting of AF. Factors with a trend for increased odds of recurrent ischaemic events were history of ischaemic stroke or TIA and large artery stenosis causing 50% or more arterial narrowing supplying the territory of the infarct. Important predictors of d-sICH were early haemorrhagic transformation, male sex and known history of hyperlipidaemia.

Mechanisms of associations

Several studies have shown that ipsilateral arterial stenosis is an important predictor of early recurrence in non-cardioembolic stroke via several mechanisms including artery to artery embolisation and infarct growth in the setting of branch atheromatous disease or hypoperfusion.10–13 Although this association has not been investigated in patients with ischaemic stroke in the setting of AF, it is likely that in some of our patients, particularly those with ipsilateral atherosclerosis causing 50% or more luminal narrowing, the mechanism of stroke may have been related to large artery atherosclerotic disease as opposed to AF, which among stroke subtypes, portends a higher risk of early recurrence.10 This is likely a specific predictor of recurrent ischaemic stroke or TIA as shown in our study. In addition, in patients with AF, prior stroke or TIA is the most important predictor of long-term ischaemic stroke risk.5 14 These studies, however, did not investigate the effect of prior stroke on early recurrence, and therefore, more studies are needed to confirm this association. Furthermore, in our study, the CHA2DS2-Vasc score was not an important predictor of early recurrence. This observation is in line with prior studies in AF6 as well as in non-AF patients where the risk of early recurrence is likely related to the specific underlying stroke mechanism, which may not be adequately captured by clinical scores that are based on the cumulative vascular risk factors and that do not consider the quality of stroke prevention measures in affected patients.10 Furthermore, in our study serum and echocardiographic factors were not associated with early ischaemic stroke recurrence. For instance, unlike prior observations,6 severe left atrial enlargement and spontaneous echocardiographic contrast on echocardiography did not portend an increased risk of early recurrence in our study. Although this may be the case, it is also possible that echocardiographic findings may have led treating physicians to pursue more aggressive strategies leading to a reduction in early recurrence risk. In addition, transthoracic echocardiography, the main modality used to assess cardiac structure and function in our study, is limited in the assessment of left atrial and left atrial appendage pathology. Thus, we may have underestimated the presence of such atrial pathology and this finding needs further study.

The second important observation from our study was the identification of several predictors of d-sICH. The identified risk factors have been shown to relate to several, not mutually exclusive mechanisms implicated in the formation of sICH, including early disruption of the blood brain barrier, coagulopathy and reperfusion injury.7 8 Specifically, markers of blood brain barrier dysfunction such as early haemorrhagic transformation were important predictors of d-sICH. In addition, the association between history of hyperlipidaemia and d-sICH is noteworthy. In our cohort, patients with a history of hyperlipidaemia were more likely to be on statin therapy and had significantly lower LDL levels than those without a history of hyperlipidaemia. Therefore, it is possible that the association between history of hyperlipidaemia and d-sICH may be at least partially mediated by statin use. Randomised controlled trials showed that statin therapy was associated with increased risk of ICH15 16 and some hypothesise that it may be related to effect of statins on the integrity of the neurovascular unit, predisposing to ‘leakage’ and posing increased risk of haemorrhagic complications.17 In patients with ischaemic stroke receiving alteplase, there is mixed evidence on the short-term benefit of statins. For instance, while one study shows that pretreatment with statins was not associated with increased haemorrhagic complications,18–20 others show increased risk of sICH with statin therapy,21 particularly when higher doses are used.22 It is, therefore, possible that while in most patients statin use may provide both short-term and long-term benefit,23 it may increase the risk of haemorrhagic complications in a subset of patients and in these patients, the benefit of high intensity therapy should be weighed against the risk of haemorrhagic complications. Furthermore, in our study, male sex was associated with increased haemorrhagic complications. Prior studies showed that the rate of sICH with alteplase is higher in men but larger studies are needed to confirm this association.24 To note that we did not include NIHSS score in the adjusted model because it correlated with ischaemic stroke volume which was included in the Cox regression model.

Clinical implications

Our study has several clinical implications. First, it is reassuring to see that the risk of early recurrence and d-sICH in patients with AF-related ischaemic stroke was fairly low. In fact, our risk estimates were lower than what has been previously reported,2 3 which provides evidence on the importance of current stroke prevention strategies to reduce this risk. Second, predictors of early recurrent ischaemic events were different from those of d-sICH, which is important for clinicians as those at high risk of recurrent ischaemic event are not necessarily at a higher risk of d-sICH. Therefore, identifying these predictors would help clinicians balance the risk of recurrent stroke against symptomatic haemorrhage when deciding on recurrent stroke versus d-sICH prevention strategies. Third, the association between ipsilateral arterial stenosis with early recurrence risk highlights the importance of vascular imaging in patients with AF to determine potential competing mechanism that portends a high risk of early recurrence. This is particularly important in patients with cervical ICA stenosis, a group who may benefit from carotid revascularisation25 and in those with intracranial stenosis who may benefit from antiplatelet therapy and high intensity stating therapy.25–27 While some studies show that echocardiography in patients with known mechanism may be useful to diagnose clinically covert cardiovascular disease,28 in our study the lack of association between echocardiographic variables and early recurrence challenges the role of routine echocardiography use to risk stratify patients with ischaemic stroke and known AF, which is line with current guidelines.4 Finally, the association between history of hyperlipidaemia and d-sICH and the possibility of this to be at least partially mediated by statins warrants further investigation, particularly in those at higher risk of d-sICH and low LDL levels.

Strengths and limitations

Our study has several limitations including its retrospective nature and involving patients treated at comprehensive stroke centres, potentially introducing bias that may reduce generalisability of our findings to smaller community hospitals. In addition, we lack data on several factors including treatments such as carotid intervention and medication dosages and medication adherence which may influence stroke and haemorrhage risk. Furthermore, imaging and echocardiographic were performed for clinical use and techniques may have varied across centres which may have added heterogeneity. Moreover, the small number of d-sICH outcomes reduced the power of our statistical analyses. Finally, since our study was not randomised, the usual clinical practice could impact the risk of ischaemic or haemorrhagic recurrence; for example, delaying anticoagulation in patients with haemorrhagic transformation. Major strengths or our study relate to the inclusion of a large, well-characterised patient cohort with the use of real world data, including pertinent clinical, laboratory, imaging and echocardiographic variables not considered in previous studies. In addition, our study included patients treated across eight centres in the USA with variability in practice and therefore may offer a good representation of the real world practices when treating ischaemic stroke in the setting of AF.

Conclusion

In patients with ischaemic stroke in the setting of AF, predictors of d-sICH differ from those of recurrent ischaemic events. Identifying and differentially considering these predictors may aid risk stratification of patients and inform early stroke and d-sICH prevention strategies. Due to our study limitations, larger prospective studies are needed to confirm our findings.

References

Footnotes

  • Twitter @neurogiles

  • Contributors SY: drafting manuscript, study concept and design, and data collection. NH, JG, ALL, EM, KF, CRLG, MK, AN and SK: study concept and design, site data management, and manuscript revision. ES and TT: Data management and manuscript revision. AL, MN, AK, IA, BMG, HF, KBE, SDA, HP, HM, JT, MV, CE, NC, KM, IM-N and MS: Data collection and manuscript revision.

  • Funding NH is supported by K08NS091499 from the National Institute of Neurological Disorders and Stroke of the National Institutes of Health and R44NS076272 from the Eunice Kennedy Shriver National Institute of Child Health and Human Development of the National Institutes of Health. ALL is supported by K23NS107643 from the National Institute of Neurological Disorders and Stroke of the National Institutes of Health.

  • Disclaimer The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

  • Competing interests SY’s department received funding from Medtronic for his work in outcome adjudication for the Stroke AF trial.

  • Patient consent for publication Not required.

  • Ethics approval Institutional Review Board approval was obtained from each of the participating centres.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data availability statement Data are available on reasonable request. Data are available for sharing on email request to corresponding author.