Michelle Christina Johansen, MD
Scheitz JF, Erdur H, Haeusler KG, Audebert HJ, Roser M, Laufs U, et al. Insular Cortex Lesions, Cardiac Troponin, and Detection of Previously Unknown Atrial Fibrillation in Acute Ischemic Stroke: Insights From the Troponin Elevation in Acute Ischemic Stroke Study. Stroke. 2015
Atrial fibrillation (Afib) is a well-documented risk factor for stroke and conversely a history of stroke in the setting of Afib yields a CHADS2 score of 2, necessitating initiation of anticoagulation.
With such a well-established causal relationship, attention has turned to diagnosis of paroxysmal atrial fibrillation with the thought that many cryptogenic strokes are falsely labeled and would be cardioembolic with prolonged monitoring. But in an era where quality of care and cost reduction go hand in hand, how far do physicians need to go in order to diagnose Afib? Should we start with a Holter monitor and if so, how long should the device be worn? What about an implantable device recorder? Is the cost justifiable?
Scheitz et al. take the paroxysmal afib discussion in a different direction through the TRELAS (TRoponin ELevation in Acute ischemic Stroke) study. Their aim was to identify predictors of previously undetected Afib discovered during in-hospital ECG monitoring. They prospectively registered patients with image confirmed acute ischemic stroke (within 72hrs) who were admitted to a tertiary care hospital in Berlin during a two year window. Exclusions were that the patient had never received a diagnosis of Afib and had no evidence of an ST elevation MI, a pacemaker or endocarditis and they had to be ECG monitored for ≥24 hours. At the time of admission labs to include a highly sensitive cardiac troponin T (hs-cTnT) were drawn and a CHADS2 score and CHA2DS2-VASc score was calculated from baseline covariates independent of whether the patient had atrial fibrillation. Imaging was evaluated to assess for the presence of insular cortex involvement as well as the presence of multiple vascular territory lesions.
The authors registered 1823 patients and analysis was performed on 1228 patients. 114 patients had paroxysmal afib detected during their in hospital stay (median recording time 3 days). The authors performed a multivariable regression analysis in order to probe for variables suitable for an Afib prediction score. In the final analysis, they found that insular cortex strokes as well as higher hs-cTnT were significantly associated with paroxysmal Afib. Older age and history of hypertension was also associated with newly detected cases.
Interestingly, the former two variable remained independently associated when excluding patients with monitoring less than 1 day or greater than 5 days. When these two variables were added to a regression model, there was significant improvement in the C statistics (decreased variability) as well as improved predictive value compared to a model including the CHADS2-score or CHA2DS2-VASc-score alone.
The issue tackled by the investigational team is an important one and represents a nearly daily conundrum for the practicing neurologist. As physicians we long for a way to predict who might need prolonged monitoring. So are we to assume then that patients with an unknown etiology and insular strokes need implantable devices? This seems too vague a definition to as yet be of any clinical utility. What about the role of the hs-cTnT? This marker first entered the cardiology community in 2009 and holds great promise in facilitating diagnosis of MI. There has been literature suggesting that higher levels of hs-cTnT has prognostic information indicating a worse mortality and morbidity following stroke. On the other hand, levels are strongly influenced by age and existing structural heart disease making the applicability of this test in these subgroups limited. Regardless, this study is certainly a step in the right direction as physicians desperately need clear guidance on how to respond to cryptogenic stroke, particularly with our aging population and increased possibility of paroxysmal Afib.