As healthcare delivery in the USA transforms into a model that at its core requires value-based considerations, ischemic stroke is confronted by intersecting forces. Modern techniques allow rapid revascularization in the majority of patients with large vessel occlusions. Dramatic advances in the evidentiary basis for mechanical embolectomy are increasing the number of patients treated with this therapy. A key part of the therapeutic arsenal in many patients treated with interventional techniques has been concurrent intravenous thrombolysis. We consider whether this paradigm warrants change.
Statistics from Altmetric.com
Endovascular stroke treatment is surging forward in a dramatic fashion. For many years the technology available was unable to match the motivation of neurointerventional (NI) specialists to treat this condition. Despite encouraging results from early randomized controlled trials (RCTs) using chemical intra-arterial (IA) thrombolysis, concerns persisted regarding hemorrhagic complications.1 ,2 As a result, there was rapid adoption of mechanical embolectomy tools as they became commercially available including the MERCI and early generation Penumbra aspiration catheters. These devices were used in several RCTs designed to evaluate endovascular stroke treatment3 but failed to show benefit, leading to questions about the efficacy of performing these procedures. The development of stent-retrievers4 ,5 and large-bore trackable aspiration catheters6 set the stage for profound improvements in recanalization rates and reduction in treatment times. The use of these new stent-retriever devices in combination with imaging-based patient selection and rapid workflow resulted in the recent landmark trials,7–11 confirming the clear benefit of embolectomy for stroke with emergent large vessel occlusion (ELVO).12 These trials changed the landscape of acute stroke treatment forever.
In the inaugural volume of JNIS our group explored the number of endovascular procedures that were occurring at that time and proposed ranges of potential cases going forward.13 Cloft et al,14 using different methodologies, proposed similar numbers at the top of their range. Most recently, Fiorella et al15 estimated in 2012 the total number of cases to be at most approximately 12 000. The evidence in support of endovascular stroke treatment for ELVO will no doubt lead to a substantial increase in the number of endovascular procedures performed. This change occurs at a time in the USA in which value-based approaches to healthcare are becoming more prominent.16 ,17 Expensive procedures or medications are a ready target for organizations aiming to control overall costs. Ischemic stroke is already under evaluation,18 and healthcare organizations will become increasingly accountable to demonstrate cost-benefit and net value. Earlier this year the Medicare and Chip Reauthorization Act delineated clear pathways forward for alternative payment models.19 Stroke neurologists and NI specialists are uniquely positioned to demonstrate the cost-benefit and net value of embolectomy for ELVO.20 Value is the relationship between input cost and the result; for the scenario of ELVO, that result is overall patient outcome. One of the many important questions to consider is the role of intravenous tissue plasminogen activator (IV tPA) in patients eligible for both IV tPA and rapid embolectomy (a bridging approach) and whether use of IV tPA in these patients is necessary and represents good value.
IV tPA: role in patients eligible for embolectomy
While there is no doubt that administration of IV tPA is the current standard of care up to 4.5 h from symptom onset, there are several considerations for and against administering IV tPA in patients with ELVO who are also eligible for embolectomy (table 1).
The follow considerations favor administering IV tPA:
IV tPA may produce recanalization and reperfusion, avoiding altogether the need for embolectomy. In the recent landmark trials, IV tPA resulted in reperfusion prior to embolectomy in 10% of patients.8 ,11 This rate is even higher in patients with a M2 middle cerebral artery (MCA) occlusion8 ,21 and could reach as high as 70% for a selected cohort of patients with residual antegrade flow though a short (<15 mm) occlusion.21 In addition to recanalization of the primary occlusion, the presence of IV tPA may enhance overall reperfusion achieved by accelerating lysis of smaller distal thrombus fragments.
IV tPA may aid embolectomy, therefore fewer stent-retriever passes may be required if embolectomy is performed after IV tPA.22
IV tPA may prevent downstream microvascular thrombosis, helping to preserve tissue viability. In an animal model, IV tPA reduced microvascular thrombosis and promoted cerebral perfusion distal to a MCA occlusion.23
Embolectomy may be delayed or simply not be feasible in a small cohort of patients. In the Multicenter Randomized Clinical Trial of Endovascular Treatment for Acute Ischemic Stroke in the Netherlands (MR CLEAN) trial, 6% of patients did not receive embolectomy because of procedural access/technical reasons.7 In these patients, IV tPA offers a chance of reperfusion if embolectomy is significantly delayed or aborted.
Considerations against administering IV tPA:
IV tPA is ineffective in the majority of patients with ELVO. Patients with ELVO from internal carotid artery and M1 middle MCA occlusions treated with IV tPA have low rates of acute recanalization, ranging from 4–8% and 26–32%, respectively.24 ,25 This contrasts significantly with the overall recanalization rates of 59–88% for embolectomy for ELVO achieved in the recent landmark trials.26
The overall clinical benefit is lower and the number needed to treat (NNT) higher in patients with ELVO. For all stroke patients eligible for IV tPA, the NNT to achieve an excellent outcome (modified Rankin Scale (mRS) score 0 or 1) is 1 in 8 if administered within 3 h and 1 in 14 if administered within 3–4.5 h.27 However, given the low rate of recanalization overall in patients with ELVO, the NNT to achieve an excellent outcome for IV tPA in patients with ELVO will be significantly higher overall, particularly in the 3–4.5 h range.
Initial triage to a Primary Stroke Center (PSC) for administration of IV tPA may delay transfer to a Comprehensive Stroke Center (CSC) for embolectomy. Patients who may be eligible for both IV tPA and embolectomy who are transferred to a PSC have lower than expected embolectomy rates and increased time from symptom onset to groin puncture for embolectomy compared with those transferred directly to CSCs.28 This warrants further consideration of the role of validated pre-hospital clinical scales for ELVO during initial triage to ensure patients are transported to the appropriate treatment facility.
IV tPA prolongs the time interval from imaging to groin puncture to perform embolectomy, which could adversely impact on clinical outcome. In the Solitaire FR Thrombectomy for Acute Revascularization (STAR) trial, the use of IV tPA resulted in a 32 min mean delay from baseline imaging to groin puncture; each 1 h increase in stroke onset to final reperfusion time decreased the odds of a good clinical outcome by 38%.29 This delay to embolectomy incurred by IV tPA is particularly notable given that the trial was restricted to high-volume CSCs; delays would likely be substantially longer in lower volume sites.30
Embolectomy may be completed and reperfusion achieved prior to completion of the 1 h long IV tPA infusion. This is not uncommon in CSCs with rapid access to embolectomy and was reported in the Endovascular Treatment for Small Core and Anterior Circulation Proximal Occlusion with Emphasis on Minimizing CT to Recanalization Times (ESCAPE) trial.9
IV tPA may increase the risk of intracranial hemorrhage. Some data suggest a trend toward higher rates of intracranial hemorrhage after IV tPA and embolectomy compared with embolectomy alone.22 Early experimental animal models revealed that IV tPA could worsen outcomes via upregulation of matrix metalloproteinases, and demonstrated that tPA is neurotoxic if it reaches the extracellular space.31 ,32 Similarly, in clinical ischemic stroke, Kidwell et al33 found that use of IV or IA tPA resulted in greater blood-brain barrier (BBB) breakdown compared with embolectomy or no treatment (p=0.002). Importantly, use of tPA was an independent predictor of BBB breakdown (p=0.001), and BBB breakdown was an independent predictor of hemorrhagic transformation (p=0.007). However, in the recent landmark trials there was no difference in the rates of symptomatic intracranial hemorrhage between the intervention or control arms,7–11 suggesting that the impact of this is probably small.
Ultimately, the main rationale for use of IV tPA in patients eligible for embolectomy is based on the argument that it remains the standard of care. While the question of clinical benefit of bridging therapy versus embolectomy alone in patients eligible for both IV tPA and embolectomy can only be answered by a RCT, it is useful to consider the cost-utility and overall healthcare value of using IV tPA in this cohort of patients.
Cost-utility and value
When considering a value-based paradigm, the overall goal of healthcare delivery is to provide superior patient value. The cost-utility of an intervention (in this case embolectomy for stroke) is the ratio between the cost of a health-related intervention and the benefit, usually measured in terms of patient-centered outcomes (eg, return of functional independence or the number of years lived in full health by the beneficiaries). This can help identify procedures and/or medications that provide the greatest benefit to patients for the lowest cost to the healthcare system. The distinction between cost-effectiveness and cost-utility is that, while both use cost in their numerator, the former uses a unidimensional measure of effectiveness such as life-years gained while the latter uses a particular measure of effectiveness that includes both quality and quantity of life (ie, the quality-adjusted life year or QALY). Contributions to the stroke literature from NI specialists have seen some efforts at this type of analysis, including with regard to ELVO.34 ,35
Recently, Ganesalingam et al36 published a cost-utility analysis of IV tPA alone compared with IV-tPA followed by mechanical embolectomy using stent-retrievers. In their model, patients who presented with an acute ischemic stroke were treated with IV tPA alone or bridging therapy (IV tPA in addition to embolectomy). They then subdivided each group into patients with a mRS score at 90 days of 0–2, 3–5 or 6, with probabilities of each outcome based upon recent landmark trial data. They studied the estimated costs and outcomes over a lifetime horizon of 20 years, with total QALYs and costs calculated for each theoretical state. In their analysis, the incremental cost of treating a patient with bridging therapy versus IV tPA alone was $12 262 per patient using a very conservative cost of approximately $3000 for the IV tPA. Patients gained an average of 1.05 QALYs/year over the theoretical 20 years of the study. The incremental cost-effectiveness ratio of embolectomy and IV tPA compared with IV tPA alone was $11 651 per QALY gained. The cost-effectiveness acceptability curves showed that embolectomy had a 100% probability of being cost-effective in their model. Notably, the assumptions made by Ganesalingam et al are based on the care model in the UK. The National Health Service (NHS) in England has mechanisms established to assess cost-effectiveness for expenditures by the NHS to decide the best value for money spent in healthcare. They use the QALY to assess the utility of the treatment. When combined with the relative cost of treatments, the data facilitate comparisons and hence decision-making for a possible expenditure versus current resource allocation by the NHS.
However, in the USA the acquisition cost, physician cost, preparation and delivery cost of IV tPA is likely to be higher than in the NHS—for example, the acquisition cost of a 100 mg vial of IV tPA at a large hospital in the north-eastern USA is $7000 compared with $3000 in the study by Ganesalingam et al. The effect of this on the overall cost-benefit of the bridging strategy in the USA remains to be seen, but we anticipate that embolectomy would remain cost-effective. Unfortunately, the Patient Centered Outcomes Research Institute (PCORI) ushered in by the Affordable Care Act is not charged with evaluating cost-effectiveness.37 ,38 Thus, the burden is upon stroke neurologists and NI specialists who treat patients with ELVO to examine this issue further.
This assessment could be simplified by considering the healthcare value equation suggested by Michael Porter from Harvard Business School.39 If healthcare value is defined as outcome divided by cost, it is clear that marked improvement in patient outcomes and overall reduced rehabilitation and nursing home cost have improved the value of performing mechanical embolectomy. Thus, for patients who present directly to a CSC who are able to access embolectomy rapidly, perhaps eliminating the additional cost and potential harms of IV tPA will further enhance healthcare value. This is a realistic prospect for CSCs that have efficient workflow processes. In the Stent-retriever Thrombectomy after IV t-PA vs t-PA alone in Stroke (SWIFT-PRIME) trial, the median time from arrival to groin puncture for embolectomy was 90 min (IQR 69–120 min).11 This overlaps with the median door-to-needle time for IV tPA of 67 min (IQR 51–87 min) from a recent analysis of participating hospitals in ‘Get With The Guidelines—Stroke’.40 To replicate the results of these landmark trials, there is renewed focus on quality improvement initiatives that result in stroke treatment workflow optimization to further reduce treatment delay.41 ,42 As workflow optimization continues to improve, routine median door-to-groin puncture times within 60 min are not inconceivable and are already being achieved at optimized CSCs using parallel workflow. This sets the stage for a RCT of IV tPA and embolectomy versus embolectomy alone for patients with ELVO. There will be debate about the various elements of RCT design such as using the non-inferiority versus superiority approach, imaging investigation, randomization protocol and other parameters. However, the goal of our group is to initiate the discussion for a RCT, not design the protocol. If such a RCT was performed and its results were convincing, data-driven stroke practitioners will adapt their practice.
With increasing pressures to optimize value and thus rein in healthcare spending, it is imperative that our established and long-standing therapies are carefully examined through the lens of value-based care, particularly with the advent and evolution of newer and more effective therapies. Until recently, IV tPA has been the mainstay of therapy for patients with acute ischemic stroke, including for ELVO. Perhaps in this new era of rapid and predictable recanalization by embolectomy, IV tPA may be an unnecessary expense to the healthcare system in a subset of patients with ELVO. Although a RCT is required to answer this question, it is certainly possible that, for patients eligible for both IV tPA and embolectomy who present to CSCs with efficient workflow processes, there may be superior patient value and overall healthcare benefit if the thousands of dollars (and the precious minutes) spent on IV tPA delivery were saved.
RVC and TML-M contributed equally.
Competing interests None declared.
Provenance and peer review Commissioned; internally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.