Stroke Alert March 2021

On Episode 2 of the Stroke Alert podcast, host Dr. Negar Asdaghi highlights two featured articles from the March 2021 issue of Stroke. This episode also features a conversation with Dr. Joan Montaner from Neurovascular Research Laboratory at the Universitat Autònoma in Barcelona, Spain, to discuss his article “D-Dimer as Predictor of Large Vessel Occlusion in Acute Ischemic Stroke.”

Dr. Negar Asdaghi:           Can your microRNA profile predict your future risk of stroke?

Is stroke that wake-up call to finally live a healthier lifestyle, better diet, exercise more, and stop smoking?

Can a simple blood test improve our clinical predictive models for presence of a large vessel occlusion in patients with suspected ischemic stroke?

We have the answers and much more in today's podcast. You're listening to Stroke Alert. Stay with us.

Dr. Negar Asdaghi:           From the Editorial Board of Stroke, welcome to the Stroke Alert Podcast. My name is Negar Asdaghi. I'm an Associate Professor of Neurology at the University of Miami, Miller School of Medicine, and the host of the monthly Stroke Alert Podcast. In today's podcast, I'm going to interview the senior author of the study on the values of D-dimer and predicting the presence of large vessel occlusion in stroke. But first with these two articles.

Dr. Negar Asdaghi:           DNA noncoding sequences and introns, once thought to represent the, quote, junk DNA, quote, have been found to play an important role in the modulation of gene expression at the post transcriptional level through coding for regulatory molecules, such as microRNAs, or miRNA. Whether the presence of certain miRNAs can signal a future risk of development of stroke is unknown. In their paper titled “Circulatory MicroRNAs as Potential Biomarkers for Stroke Risk: The Rotterdam Study,” Dr. Michelle Mens and colleagues from the Department of Neurology, University Medical Center, in Rotterdam, Netherlands, discuss their findings related to microRNA samples collected between 2002 and 2005 from over 1900 stroke-free participants of the Rotterdam Study. Participants were assessed for incident stroke through continuous monitoring of medical records until January 1, 2016.

Dr. Negar Asdaghi:           At baseline, using next-generation sequencing, they measured expression levels of over 2083 miRNAs in plasma samples. During a mean follow-up of close to 10 years, the incidence of stroke was 7% in their study population, and they found, in total, 39 miRNAs were at least nominally related with that incidence of stroke. In their fully adjusted model, they found significant association between expression level of three particular microRNAs and risk of stroke, with the hazard ratio ranging between 1.1 to 2.6. Interestingly, the area under the curve for the longitudinal predictive models improved when the miRNA data was added to the vascular risk factor model. And in conclusion, they found miRNA 6124, 5196-5p and 4292 were associated with future risk of stroke in their population. The elevated levels of these miRNAs may serve as plasma biomarkers for predicting future risk of stroke in combination with other known vascular risk factors for stroke.

Dr. Negar Asdaghi:           So, speaking of vascular risk factors, let's move on to our second paper for today's podcast. There's a growing emphasis on adherence with pharmaceutical interventions, such as diabetic and blood pressure treatments, statin therapy, to control the risk factors for stroke and prevent recurrent vascular events. All the while, the non-pharmaceutical interventions, such as smoking cessation, diet control, and increased physical activity, seem to represent the somewhat easy or implied aspect of our secondary preventive efforts. But how well are stroke survivors doing with regards to making these healthy lifestyle modifications? In the March issue of Stroke, Dr. Chelsea Liu and colleagues from Johns Hopkins School of Public Health presented their findings on lifestyle and behavioral changes pertaining to cardiovascular health in the study titled, “Change in Life’s Simple 7 Measure of Cardiovascular Health After Incident Stroke: The REGARDS Study.”

Dr. Negar Asdaghi:           So, this was a population-based, epidemiological study of over 7,000 stroke-free participants between 2003 and 2007, who had data on Life's Simple 7, what the author called “LS7 measures,” which studied seven different domains. Four of them behavioral, including smoking, diet, physical activity, body mass index, and three medication-controlled, including blood pressure, total cholesterol, and fasting glucose, both at study entry and their follow-up visit. At which point, either they did not have a stroke or had an ischemic stroke and were included if that stroke had happened more than one year before the follow-up visit. And so the study authors hypothesized that those with a stroke would have had a significant improvement in their Life's Simple 7 data poststroke as compared to the stroke-free participants.

Dr. Negar Asdaghi:           But what they found was completely the opposite. At 10 years follow-up, a total of 149 patients had suffered a stroke in their study. On a scale of zero to 14 at study entry, all participants scored low or relatively low in these seven simple measures, but those participants who would ultimately suffer a stroke scored significantly lower at baseline. What was alarming, though, was that after adjusting for all confounders, at follow-up, participants who had experienced an ischemic stroke showed a significantly further decline in their total LS7 score at 10-year follow-up. And the greatest declines were noted in behavioral domains, most notably physical activity and diet scores. The authors noted a non-significant improvement, in other words, improvement in weight in the BMI score among stroke survivors, but they caution that that may indeed be actually related to muscle loss, a downstream effect of decreased physical activity poststroke, rather than representing active dietary interventions with weight loss. So, in summary, this important paper highlights, on a population level, the urgent need for behavioral interventions to improve secondary prevention after a stroke event up and beyond our efforts to improve medication adherence.

Dr. Negar Asdaghi:           So now moving on from secondary preventative measures to the acute phase, our next paper discusses ways in which we can improve our diagnostic accuracy in the acute setting. Identification of large vessel occlusions is the first step in determining patients' eligibility for endovascular thrombectomy, a highly effective treatment to improve outcomes in acute ischemic stroke. But without vascular imaging, which may not be readily available in the small or community hospitals, the decision to transfer patients to thrombectomy-capable centers is entirely dependent on clinical scales, which, as we all know, may have suboptimal sensitivity and specificity. So the question is, could a simple blood test improve the predictive capabilities of our current clinical scales for presence of a target LVO, or large vessel occlusion? Joining me now is Dr. Joan Montaner from Neurovascular Research Laboratory at the Universitat Autònoma in Barcelona, who is the senior author of the study titled “D-Dimer as Predictor of Large Vessel Occlusion in Acute Ischemic Stroke.” Good morning, Joan, all the way from the sunny Florida to the beautiful Barcelona. Good to have you with us, and thank you for joining us.

Dr. Joan Montaner:         Hello. Nice to talk with you on blood biomarkers for stroke management.

Dr. Negar Asdaghi:           Thank you, Joan. Your study touches on the importance of improving the ways in which the systems of care are set up in triage and transfer of patients with thrombectomy-capable centers. Can you please tell us briefly about the stroke systems of care in Catalonia where you practice and where your study is based out of? And what clinical scales are currently used for transfer of patients with suspected acute stroke to a comprehensive stroke center?

Dr. Joan Montaner:         Yes, Catalonia, it's a region of about 7.5 million inhabitants. And when we did this study, most of the comprehensive stroke centers were located in Barcelona itself, in the capital. So it's true that there are several areas of the region that are far away from Barcelona. It took more than two hours to bring some patients from those distant regions to Barcelona. That's why we began to use these clinical scales that you are talking about. Mainly they are RACE, it's like a simplification of the NIHSS subscale. And, in fact, a large study RACE card that was presented last year in the European Stroke Conference was done to try to see if we could, by using these scales, RACE, select the right patients to come directly to the thrombectomy centers instead of going to the closest hospital. But, unfortunately, the results were neutral. So, we were a little bit disappointed, and we think, as you were saying, that these neurological scales are suboptimal, probably not enough sensitivity and specificity for identifying LVO. That's why we think that these biomarkers could improve the accuracy of those scales.

Dr. Negar Asdaghi:           Perfect. I totally agree with you. And now, before you tell us about the biomarkers, can you just briefly tell us about the Stroke-Chip study, your study population, and what prompted you to look at these various biomarkers that you addressed in the paper?

Dr. Joan Montaner:         Stroke-Chip was a lot, it was really a massive collaborative effort among all the public hospitals in this network here in Catalonia. We were able to collect more than 1,300 patients in this particular study that we are talking about. Dr. Anna Ramos-Pachón and Elena Cancio were leading the analysis on the relation of these biomarkers with LVO. But I have to say that this was not the original intention of our study. Really, and perhaps we were naive at that time, we were looking for biomarkers to differentiate ischemia from hemorrhagic strokes or from stroke mimicking conditions to try to give TPA or TNK in the ambulance. But, as I was saying, perhaps that was a little bit naive, and we know how difficult that would be and perhaps with some liabilities. That's why it came this idea of, "Well, if we use those markers, not for giving a drug in the ambulance, but for doing triage and sending the patient to the right hospital, that could be more simple and more useful even."

Dr. Negar Asdaghi:           Thank you very much. Can you briefly tell us about the study? What were your inclusion criteria?

Dr. Joan Montaner:         Well, in this study, we selected all consecutive acute stroke patients attending the stroke unit of all these hospitals. We were including all stroke suspicions, if their symptoms onset happened within six hours. So, it's really hyperacute patients. And we were able to collect, like this, more than 1,300 patients. And then at the hospital, with the angio CT or duplex, we were able to categorize those with LVO, and we measured a panel of different biomarkers in the blood stream of those patients and trying to associate which of these markers were related with having or not having an LVO.

Dr. Negar Asdaghi:           Very interesting. So tell us, please, your study’s main finding?

Dr. Joan Montaner:         The main finding, what we liked more, let's say, of our results was that some of those markers, specifically NT-proBNP and D-dimer, were really high among patients with a large vessel occlusion. When we combined these results, for example, having high levels of D-dimer, those patients above fourth quartile of D-dimer with more stroke severity, patients with NIH of more than 10, the accuracy was really good. It was very specific, 93% specificity, 34% sensitivity, to predict an LVO. So this means that without almost any mistake, you select more than one third of the patients that have an LVO, that could be very useful. To bring those patients, we were talking from far away of these thrombectomy centers, to the right place. And perhaps we could be doing a thrombectomy one or two hours before with these technologies.

Dr. Negar Asdaghi:           Perfect. So basically, just to reiterate what you're saying, is that D-dimer, as non-specific as it is and as important as it is to note that it can be elevated in the setting of aging or increase NIH Stroke Scale severity, this increase in D-dimer noted in patients with LVO was just not a factor of just age simply or increased severity of the stroke scale. Can you tell us about your multivariate analysis and what other factors you adjusted for in your final model?

Dr. Joan Montaner:         You are right that D-dimer can be modified by many things, as you were saying. That's why we took a lot of care about the multivariate analysis and all factors, all clinical factors that were related with LVO were included in the model. And finally, only eight NIH Stroke Scale scores D-dimer and the vast history of atrial fibrillation were included in the model. Odds ratio for D-dimer was 1.59 that I think it's quite acceptable. And it's true that in that model, NT-proBNP was not included anymore, probably because it's related with a fee. So, that's something interesting if perhaps in the ambulance, you don't know about the story, the history of a patient, of a fee, we could use NT-proBNP, so I think this opens the possibility of using different clinical neurological scales biomarkers in combination to make the prediction of LVO.

Dr. Negar Asdaghi:           Yes. Very, very exciting results for sure. So what is our main takeaway from your study? Are we thinking that D-dimer or a particular level of elevations of D-dimer will one day become the, quote, Troponin equivalent of LVO for stroke?

Dr. Joan Montaner:         Well, it sounds nice, but I know it's several technical issues here. You are right that there is variability among labs in the measurement of D-dimer so now what we are doing is really, in a prospective study called BIO-FAST in the south of Spain, in Seville, in a large network of ambulances, we are measuring D-dimer, but in a rapid fashion with a rapid point of care test in the ambulance itself. We think that we are not going to have a magic biomarker. Not that Troponin you are talking about. Probably we need to combine it with others. We think that the marker of brain damage would add a lot on top of D-dimer, probably D-dimer is very good for the clot burden, but we think other markers could improve the accuracy of the test. And we are measuring them together with these. Our dream would be really to have cost utility study in the future and to see if really we are able to randomize patients based on these biomarkers in the ambulance, will have an impact on outcome if we are able really to do thrombectomies much faster.

Dr. Negar Asdaghi:           Well, we certainly look forward to covering your future studies on this topic of biomarkers. Dr. Joan Montaner, thank you for joining us and congratulations on your work.

Dr. Joan Montaner:         Thanks a lot.

Dr. Negar Asdaghi:           And this concludes our podcast. Don’t forget to check online for the full list of publications, including two papers on the state of pediatric thrombectomy and a study on the association between stroke and subsequent risk of suicide that are published online ahead of their presentations at the International Stroke Conference. Until our next podcast, stay alert with Stroke Alert.