Editorial - do not throw the baby out with the bathwater….

On April 11th the NIH issued a Clinical Alert stating that the National Institute of Neurological Disorders and Stroke (NINDS) had stopped enrollment in a trial that compared an aggressive medical therapy alone with aggressive medical therapy combined with intracranial angioplasty and stenting. The necessity to perform such a trial essentially arose from a series of studies published between 2005 and 2007: The Warfarin-Aspirin Symptomatic Intracranial Disease (WASID) trial showed that patients with previously symptomatic intracranial stenosis had a 20% risk of recurring stroke within one year if the stenosis was larger than 70% and if they were treated with Aspirin alone 1, whereas the first Wingspan studies demonstrated stroke rates of 6 to 9% with a high rate of technical success indicating that intracranial stenting may represent a viable alternative for treating high grade intracranial stenoses 2. The Stenting and Aggressive Medical Management for Preventing Recurrent stroke in Intracranial Stenosis (SAMMPRIS) trial was the first prospective randomized multicenter trial evaluating the treatment of first-time symptomatic intracranial high-degree (>70%) stenosis. Two treatment arms were evaluated against each other: a) intensive medical therapy alone (i.e. in addition to DUAL antiplatelet therapy (aspirine and clopidogrel) aggressive risk factor management with targeting blood pressure ( 70 mg/dl), and, b) the same intensive medical therapy with additional stenting and angioplasty employing a self-expandable stent system. Patients were included after a qualifying event (TIA or Stroke) that could be attributed to a high degree intracranial stenosis 3. Enrollment was stopped after 14% of patients treated with angioplasty combined with stenting experienced a stroke or died within the first 30 days after enrollment compared with 5.8% of patients treated with medical therapy alone. There were five stroke-related deaths within 30 days after enrollment, all in the stenting arm 4. Does this mean that we should discontinue stenting for intracranial stenoses now for good? In order to answer this question one should take a second look at the SAMMPRIS trial and what is has shown – and what not. The results of the trial are surprising in both therapeutic arms: 1. The 30-day rate of stroke or death in the intensive medical treatment arm was substantially lower than estimated (based on historical controls, most of whom received standard medical care) and definitely lower than 20% as indicated from the WASID trial. This substantial drop in the annual risk for subsequent stroke under “best medical therapy” has “raised the bar” for intracranial stenting: to become a viable alternative for first-time symptomatic patients with high-grade stenoses intracranial stenting should have a risk that is lower than 6% which at present is unlikely to happen even in very experienced hands. 2. The 30-day rate of stroke and death (14%) in the stented patients is substantially higher than the estimated rate of 5% – 10% based on registry data (such as the Intrastent registry 5) or on data from high-volume centers. This discrepancy raises a series of questions that will presumably be answered by the full publication of the trial that was conducted in 50 different sites across the United States and that enrolled 451 patients: – Could the higher complication rate between these studies and the SAMMPRIS trial be related to lower complication rates in high-volume centers, a phenomenon that was observed in previous studies 6 that aimed to compare a technically demanding procedure to conservative treatment, or a highly standardized or routine procedure? A high-volume center is not automatically a high-enrollment center and vice versa. An operator in a high-volume multi-operator center may have less experience compared to single-operator sites. If there was a decrease in complication rate with increasing experience of the operator, one could make a point for centralization of certain procedures or more rigorous quality requirements for the operators. – Could the higher complication rate in SAMMPRIS be related to patient characteristics? Recent studies have demonstrated that specific risk factors can be identified that lead to a higher complication rate in intracranial stenting such as increased patient age 7, posterior circulation stenosis, and stenting soon after the qualifying event 8. Can the SAMMPRIS trial similarly identify a specific patient population that is at an unacceptable high risk and other populations that are not? – Could the higher complication rate in SAMMPRIS be related to characteristics of the stenosis and its associated pathomechanism? In high degree intracranial atherosclerosis, multiple pathomechanisms can be put forward that may explain stroke or TIAs including distal thrombembolism, hemodynamic compromise with watershed territory infarctions, and local perforator occlusions due to the intramural plaque 9. One may hypothesize that angioplasty and stenting of stenoses that became symptomatic due to local perforator occlusions may be at a higher risk of periprocedural complications due to further egression of plaque material into local perforating branches. On the other hand, it seems difficult to conceive that patients presenting with hemodynamic infarctions (i.e. truly related to a low flow situation) would benefit from medical therapy alone. The SAMMPRIS trial has to be commended for demonstrating that “best medical therapy” can significantly lower the risk of subsequent stroke following a qualifying event in intracranial high-degree stenosis. This trial has also shown that intracranial stenting is a risky procedure. Therefore, in first-time symptomatic patients with intracranial stenosis medical therapy (including the aggressive treatment outlined in the trial) should be the therapy of first choice. The trial, on the other hand, cannot give recommendations about the “medical failures”, i.e. those patients who are repeatedly symptomatic despite the best medical treatment, which may be related to a variety of factors including hemodynamic compromise related to a high-degree stenosis without sufficient collaterals. In many centers including ours patients are only considered for intracranial stenting if they have failed best medical therapy. If the SAMMPRIS trial were to lead to a change in this paradigm or if we were to deduce from this trial that stenting has no role in the management of repeatedly symptomatic intracranial stenotic disease, we would “throw out the baby with the bathwater”.