Letter by D’Anna et al Regarding Article, "Long-Term Mortality in Patients With Stroke of Undetermined Etiology"
To the Editor:
We interestingly read the article Long-term mortality in patients with stroke of undetermined etiology by Nam et al,1 published in the August issue.
The authors retrospectively analyzed mortality and short-term functional outcome of 3278 patients with first ever ischemic stroke registered in the Yonsei Stroke Registry over a 10-year period (July 1997–June 2007). They found that the mortality of patients with ischemic stroke subtype with incomplete evaluation was second only to cardioembolism.
We want to congratulate with Dr Nam and his colleagues1 for the results, but we would like to bring 3 issues to the authors’ attention.
First, we would like to discuss a methodological aspect. The authors defined stroke with incomplete evaluation as the subtype in which essential studies such as brain imaging (computerized tomography/MRI), angiographic evaluation, or ECG were not performed. Echocardiography was not considered as an essential study. This is in contrast with Adams et al,2 cited by the authors. In fact Trial of Org 10172 in Acute Stroke Treatment (TOAST) classification system is based on many tests, including cardiac imaging. Like many clinicians dealing with ischemic stroke, we consider echocardiography study as a clinical routine examination for detection of cardiac sources of embolism, complementary to ECG. In addition, the authors declared that 13.5% of their patients with cardioembolic stroke did not undergo a cerebral angiographic study (computerized tomography angiography, magnetic resonance angiography, or digital subtraction angiography). It is not clear why in this 13.5% the authors did not perform at least an ultrasound study. To increase preciseness and reduce the likelihood of misclassification of patients in the other categories, Adams et al2 recommended not to classify study patients in a defined category (the cardioembolic one in this case) without performing essential diagnostic tests. In fact it is possible to define a probable diagnosis of cardioembolic stroke only when potential large artery atherosclerotic sources of thrombosis or embolism have been eliminated. This rigorous method allows the reduction of interobserver variability; whereas the application of an individual definition of incomplete evaluation could increase variability, as acknowledged also by the authors.
In the Discussion, the authors declared that there are 3 main explanations to justify high mortality in patients with an incomplete evaluation: missing of angiographic evaluation, severe neurological deficits and lower percentage of patients treated with thrombolysis, antithrombotic agents, and statins. However, individual studies3 and metanalysis4 show that admission to an organized stroke care is recommended for all patients with acute stroke, mainly because of its positive effects on the outcome and its specialized and multidisciplinary diagnostic approach. For this reason we suggest that the setting of care should be considered as a confounding factor and that it should be discussed whether higher mortality in patients with incomplete evaluation is related to the type of ward at admission, a kind of analysis that is lacking in this study. It is possible, in fact, that undetermined strokes could be more frequent in nonspecialized units.
The authors used multivariate analysis when comparing every category of stroke with small vessel artery disease and considered it as a perfect predictor. In our opinion, a comparison between determined and undetermined etiology could be more useful to investigate the effects of a stroke diagnostic assessment on the prognosis and to demonstrate the importance of an etiopathogenetic diagnosis according to TOAST criteria. Besides, a comparison of the outcomes among all the different TOAST categories is difficult to understand effectively, because of the unknown composition of the undetermined etiology group. This could result in increased statistical differences in some cases and a lack of statistical differences in others.
In conclusion, we suggest that a rigorously designed observational study, which includes the effects of Stroke Unit care and a direct comparison between ischemic strokes of undetermined etiology and determined etiology, is essential to understand the effects of both stroke etiological diagnosis and setting of care in everyday clinical practice.
Lucio D’Anna, MD
Francesco Janes, MD
Gian Luigi Gigli, MD
Chair of Neurology
Department of Experimental and Clinical Medical Sciences
University of Udine Medical School
Department of Neurosciences
S. Maria della Misericordia
Stroke welcomes Letters to the Editor and will publish them, if suitable, as space permits. Letters must reference a Stroke published-ahead-of-print article or an article printed within the past 3 weeks. The maximum length is 750 words including no more than 5 references and 3 authors. Please submit letters typed double-spaced. Letters may be shortened or edited. Include a completed copyright transfer agreement form (available online at http://stroke.ahajournals.org and http://submit-stroke.ahajournals.org).
- © 2012 American Heart Association, Inc.
- Nam HS,
- Kim HC,
- Kim YD,
- Lee HS,
- Kim J,
- Lee DH,
- et al
- Adams HP Jr.,
- Bendixen BH,
- Kappelle LJ,
- Biller J,
- Love BB,
- Gordon DL,
- et al
- 4.↵Stroke Unit Trialists' Collaboration. Organised inpatient (stroke unit) care for stroke. Cochrane Database Syst Rev. 2007;4: CD000197. http://www.thecochranelibrary.com.