Robert W. Regenhardt, MD, PhD
Goldhoorn R-JB, Verhagen M, Dippel DWJ, van der Lugt A, Lingsma HF, Roos YBWEM, et al. Safety and Outcome of Endovascular Treatment in Prestroke-Dependent Patients: Results From MR CLEAN Registry. Stroke. 2018
Robust, randomized trial evidence exists supporting the efficacy of thrombolysis and thrombectomy to reduce disability after acute ischemic stroke for select patients. However, there is a paucity of evidence for patients with pre-stroke disability since patients with baseline mRS 2-5 are often excluded from trials. Like many acute stroke care providers, I have found myself in several situations in which our team pauses as we learn the pre-stroke mRS is not perfect. It spurs some debate, as in many cases the pre-stroke disability seems to have little implications for hemorrhage risk, procedural risk, or treatment efficacy. Ganesh et al. point out that exclusion of patients with mRS 2-5 is not based on mechanistic hypotheses about reduced benefit for this population necessarily, but reflects that pre-stroke disability prevents patients from contributing to the typical dichotomized mRS analyses. They further discuss that patients with mRS 2-4 would likely consider retaining their pre-stroke status a favorable outcome, but pre-stroke disability is often cited as a reason for withholding treatment.
The 2018 AHA stroke guidelines do take a more aggressive stance overall on both thrombolysis and thrombectomy in several situations, including for patients with pre-stroke disability. Regarding thrombolysis, the guidelines emphasize that it may be reasonable to treat these patients; pre-stroke disability does not independently increase hemorrhage risk, but it may be associated with less improvement and higher mortality. Regarding thrombectomy, they also state that it may be reasonable to treat patients with pre-stroke disability, acknowledging the benefit is uncertain. Two articles in the October issue of Stroke add to the literature regarding the natural history of stroke in patients with pre-stroke disability and the safety and efficacy of thrombectomy in this population.
Ganesh et al. prospectively studied 1607 stroke patients from 2002 to 2014, of which 530 (33%) had a pre-stroke mRS of 2-4. Of those with pre-stroke disability, only 2 received thrombolysis, but 421 (79%) were alive after 3 months. Delta mRS was calculated as the difference between 3-month mRS and pre-stroke mRS, and increased delta mRS was associated with increased 5-year mortality/institutionalization and increased healthcare/social-care costs (see Table 2). Indeed >40% were left with further disability or requiring new institutionalization. The authors conclude that patients with pre-stroke disability have higher mortality, institutionalization, and costs if they accumulate additional disability because of stroke, urging future trials and registries to include these patients. They acknowledge that their dataset had a low median NIHSS, which could limit generalization to more severe stroke patients, and while they adjusted for age, sex, NIHSS, they did not adjust for other variables that could influence their outcomes.
In the same issue of Stroke, Goldhoorn et al examined the MR CLEAN prospective registry from March 2014 to June 2016; 1441 anterior LVO patients who underwent thrombectomy were analyzed, with 157 (11%) having pre-stroke mRS 3-5. Three-month favorable outcomes (mRS 0-2 or not worsening) was seen in 27% of patients with pre-stroke mRS 3-5 and 42% of patients with pre-stroke mRS 0-2; however, after adjustment, pre-stroke mRS 3-5 was not associated with less favorable outcome (see Table 3). Symptomatic intracerebral hemorrhage and ischemic stroke progression were not different comparing patients with pre-stroke mRS 3-5 vs. those with pre-stroke mRS 0-2. The authors conclude that a substantial proportion of patients with pre-stroke disability will reach their pre-stroke mRS after thrombectomy, and complication rates are not increased in this population; therefore, this patient population should not be routinely excluded from thrombectomy. They acknowledge several limitations, including their use of the mRS score itself for this population, given its decreased discriminative ability for neurologic deficits at higher scores, and that they didn’t analyze the underlying cause of pre-stroke disability.
The decision to treat patients with pre-stroke disability is complex and probably best viewed through the lens of the 4 basic principles of medical ethics. Regarding beneficence, growing data support that these patients stand to benefit from acute stroke treatments. Goldhoorn et al. showed that after adjustment, pre-stroke disability was not associated with less favorable outcomes after thrombectomy, and a substantial portion of patients return to their pre-stroke mRS. While an RCT is lacking, data are accumulating, and it stands to reason that these patients will benefit as long as they lack comorbidities that would otherwise limit life expectancy or prognosis. Indeed, it may be difficult to recruit patients for an RCT with mRS 2-3 given questionable equipoise.
Regarding non-maleficence, Goldhoorn et al. also demonstrated that there are no increased risks, including symptomatic hemorrhage, after thrombectomy for this patient population. Ganesh et al. showed that these patients do accumulate disability if treatment is withheld, and this worsening disability is associated with increased mortality, institutionalization, and healthcare costs.
Regarding justice, most would likely concur that patients with disability deserve the same access to care that those without disability enjoy. Perhaps autonomy deserves the most focus. Ultimately, we should involve patients and families in these difficult decisions. These decisions are too intricate and individualized to be made without them. Furthermore, we should avoid the temptation to impose our value systems. Reflecting on my own patients, I have several with mRS 4 who would absolutely want to pursue all measures to prevent additional disability.