International Stroke Conference
January 24–26
Kevin S. Attenhofer, MD
Following a mixed reception at the reveal of the new 2018 AHA/ASA Guidelines for the Early Management of Patients with Acute Ischemic Stroke, selected members of the guidelines panel spent time Thursday with conference-goers to field questions. The session was in such demand that the fire marshal had to start turning people away at the door, leading to the creation of a second session in the afternoon. All told, the panel spent about 3 hours answering questions from stroke practitioners in the trenches.
In this post, I highlight 6 of the most contentious topics brought up by clinicians. I have attempted to summarize the lengthy and passionate discussion between the panel and the crowd. The panel frequently reminded the crowd that these guidelines are here to help guide clinical practice when the clinician has a question, not dictate it. These are much more streamlined guidelines than previous versions, as the authors feel they have “trimmed the fat” to make them more usable and accessible.
The panel repeatedly stressed evidence-based practices that have a measurable, objective change in patient outcomes (morbidity, mortality, etc.) as opposed to those that simply change management without a proven outcome benefit. “Evidence-based” doesn’t mean that every guideline is based on Level A evidence, but rather that they are based on the best available evidence and the level of evidence is clearly delineated. In fact, these guidelines identify gaps where we need clinical research to determine best evidence-based care and challenge clinicians to fill the gaps.
Regarding the downgrade of the recommendation for dysphasia screening, the panel first took a moment to clarify that dysphasia screening before the patient begins eating, drinking, or taking PO medications is REASONABLE to identify patients at risk for aspiration pneumonia; however, it is not NECESSARY. They admit that this recommendation is a departure from current quality metrics. Their rationale is that the available evidence shows that dysphasia screening identified patients with dysphasia; however, there is no evidence supporting improved patient outcome as a result of this information.
Regarding the timing of carotid endarterectomy after acute ischemic stroke, the panel noted that there are now several case series that show that the risk of recurrent stroke in patients with a symptomatic carotid artery stenosis within the first 1–2 weeks is very high. At the same time, a recent meta-analysis shows high risk of complication with CEA within the first 48 hours after acute ischemic stroke. Therefore, the panel recommends CEA AFTER 48 hours, but then as soon as possible. To take this one step further, they also recommend that providers evaluate patients for CEA early (preferably within the first 24 hours) so that potential CEA patients can be identified for early surgical planning.
Regarding tenecteplase for acute ischemic stroke, the panel noted that the most recent data incorporated into the guidelines is data from the NOR-TEST study. In this trial, TNK was not superior but was very comparable to tPA. Since this trial was not designed to be a non-inferiority trial, the panel could not say that both treatments are equal. Ultimately, TNK is included in the guidelines as a IIb recommendation because the authors of the guidelines wanted to acknowledge the results that suggested similar efficacy. As you may have read by now, EXTEND IA TNK demonstrated that TNK at a lower dose of 0.25 mg/kg was not only non-inferior to tPA, but it was superior to tPA in treatment of LVO. The panel acknowledges that this can potentially change practice, but it was not available to be included in these guidelines. They hope future revisions of the guidelines may incorporate this most recent TNK study, as well as others that are ongoing.
Regarding the recommendation that EMS should triage acute stroke patients to the nearest tPA-capable center rather than bypassing for nearby thrombectomy-capable stroke centers (TCSC), the panel again acknowledged that they are in conflict with various currently available guidelines and quality measures (members of the crowd mentioned Get With the Guidelines data, as well as the Mission Lifeline Guidelines, which suggest bypass to TCSC when it is within a certain pre-defined distance). The new guideline recommends that when several tPA-capable hospitals exist within a local geographical region, they should not be bypassed for a center with a higher level of care.
Again, the panel reinforced the concept of evidence-based patient outcome data. If you bypass based on X parameters/score, the patient will do better. Does data supporting that statement exist? The panel says no. They state that as of October 2017, the false positive rate for pre-hospital stroke scales is > 50%. Many of these patients do not have a stroke. And of the ones who do, many do not need the more advanced treatment offered by the higher level of care facility. The panel also suggests that we may be doing these patients harm by driving further to a TCSC (citing an example of delayed tPA and of a status epilepticus patient mis-triaged as stroke and waiting 30 minutes longer for treatment as a result).
Regarding in-hospital institution/evaluation of secondary prevention, the authors of the guidelines felt that “early management” includes everything that happens during the hospitalization until the patient is discharged, including decisions about secondary prevention. Therefore, they included a controversial downgrade in the level of evidence for “routine stroke work-up.” They made a point to emphasize the word “routine.” In this context, “routine” is what we do in every stroke patient. Until today, every stroke patient got an MRI, Echo, risk stratification, and vessel imaging as part of a routine work-up. But is there evidence that this changes patient outcomes for the better? The panel stresses that these tests are often ordered unnecessarily and, as a result, are not cost effective and do not improve patient outcomes.
MRI was mentioned specifically by multiple members of the crowd. Panel members reminded the crowd that stroke is a clinical diagnosis and pointed to cost effectiveness analyses of “routine” MRI in patients with stroke. They feel it is not cost effective because of false positives, false negatives, and the majority of cases where the MRI does not change the patient outcome (note that I said “outcome,” not “management”).
Their point is that not every patient with a stroke needs an MRI, and that the new guidelines allow practitioners to ask the question “does this patient need an MRI?” This recommendation (as with the other recommendations not to acutely pursue secondary work-up) is followed immediately by a lower class of recommendation that the use of MRI in some patients may be needed despite the unknown outcome effect. With this, they concede that some patients will need an MRI, but it is not “routine” and should be performed at the discretion of the physician. They feel that this recommendation will also prevent insurance companies from denying tests for stroke patients based on the new guidelines; however, the crowd seemed unconvinced.
Finally, regarding pharmacological DVT prophylaxis, the new guidelines recommend IPCs over unfractionated or LMW heparins based on data from CLOTS3. There is data that demonstrates prevention of DVT and PE with heparinoids; however, this comes at a cost of increased rates of intra- and extracranial hemorrhage. Overall, the panel feels that the risks and benefits of pharmacological DVT prophylaxis balance to a net neutral effect on patient outcome, which is less favorable than the positive outcome of IPCs alone. Again, they acknowledge that they are in disagreement with current quality measures.
You may have noticed by now that times are changing. At many points, the new guidelines appear to be in conflict with quality recommendations. How do we reconcile this? Dr. Barbie Gatton from NYP Brooklyn Methodist recalled a similar situation with her ER and the discrepancies between quality measures to give beta blockers before cath lab in STEMI patients and emerging evidence reporting increased mortality with this measure. “We had to decide whether to fail our quality measures or kill patients,” she said. “Go with evidence and let the quality measures catch up.”
