Ethics

It bears repeating that as physicians, we have an obligation to offer treatments that are consistent with professional standards of care and that actually confer benefit to the patient. Failure to do so by offering futile treatments leads to subtle adverse consequences such as increasing a patient’s pain and discomfort in the final days and weeks of life, giving family false hope and delaying palliative care.

These decisions are inherently difficult and complex. Our own biases as physicians can conflict with the patient’s actual needs.3 The central question should really be, “Does the intervention have any reasonable prospect of benefiting this patient?” Age should not be the only factor. The better we understand the patient, the better we can successfully answer this question. For stroke treatment, however, there is always the urgency of time that limits our ability to accumulate sufficient information. As much as we would like to use empirical studies, our own clinical experiences, and professional consensus, taking too much time to do this hurts the patient. Methods to quickly and reliably understand prestroke morbidity that are relevant to thrombectomy are clearly needed.

But once the determination of medical futility has been made, it is important to remember to be compassionate. Showing sensitivity to patients and families is important. For example, rather than saying to the family, “there is nothing we can offer you,” it may help to emphasize that “everything possible will be done to ensure the patient’s comfort and dignity.”4

Costs

Since most thrombectomies are transferred between hospitals, the costs begin to accumulate once the decision to transfer has been made. These costs include transportation costs, staffing costs for the physician, technologist, nurse, and anesthesiologist, device costs, hospital bed and procedure room costs, spoke hospital capacity costs, and costs for the patient’s family to be nearby. Emergency medical helicopter or fixed-wing transportation costs alone can range from $12 000 to $25 000 per trip.5 Intensive care unit admission, additional diagnostic testing, and multiple consultations ensue. Turk et al calculated the costs associated with thrombectomy cases including femoral sheaths, guidewires, and catheters based on advertised manufacturer suggest retail price and found the mean cost across all groups regardless of whether a stent retriever was used was $11 926.45, with a range from $3296.00 up to $60 872.91.6

Furthermore, a single-center urban tertiary care comprehensive stroke center reported that more than half of thrombectomy transfers over a 2-year period after 2015 did not even undergo a thrombectomy.5 Lastly, it is conceivable that an excessive number of futile transfers and futile thrombectomies may contribute to career burnout among providers.7 8

Recent advances

Efficiently determining pre-stroke morbidity is an essential component in elderly stroke thrombectomy patient selection. Unfortunately, this process currently consists of often unreliable historical data, interpreted by biased family members, who for various reasons, may not be providing accurate or complete information. Trying to determine the relevance of all this information, often over the phone, in the middle of the night, and urgently, is haphazard at best.

Several recent JNIS publications, however, promise to address this gap in patient selection information by using imaging as opposed to history, as a surrogate for prestroke morbidity determination.

Leukoaraiosis has been repeatedly shown to be associated with higher rates of futile recanalization. Most recently, a single-center study evaluated whether subtracting additional points from the 10-point ASPECTS scale based on the presence of mild, moderate, or severe leukoaraiosis better predicted functional outcomes when compared with ASPECTS scoring alone. They found an area under the curve superior to that of traditional ASPECTS.9 This single-center review provides support to using white matter changes on CT to improve thrombectomy patient selection.

A single-center Australian study illustrated the role of brain atrophy in predicting futile recanalization. They used automated cerebrospinal fluid (CSF) volume calculations from baseline brain CT scans and showed an association with functional outcomes after thrombectomy.10 Using Pydicom software on plain head CTs, they used different Hounsfield unit thresholds to segment brain tissue and intracranial CSF. They then reported intracranial CSF volume as a proportion of total intracranial volume. They found for every 5% increase in CSF volume, there was a 35% lower odds of functional independence and 59% higher odds of a worse mRS score after thrombectomy.

Automated programs to evaluate brain atrophy have limitations, however. Artifacts such as encephalomalacia or cerebral edema may be difficult to account for. Widespread implementation in all emergency rooms may be a logistical challenge, likely similar to the challenges in dissemination of advanced perfusion imaging processing software to evaluate extended time window thrombectomy candidates. Post-processing software is notoriously expensive to implement and maintain, which limits scalability. To maximize accessibility, more simple and practical methods to determine brain atrophy may have value.

Even more recently, visual brain atrophy measurement was shown, in a single-center study in Finland, to independently predict 3-month mortality after stroke thrombectomy.11 Furthermore, they found that age loses its significance as a predictor of mortality when compared with brain atrophy measurement. Using a plain head CT may represent a more practical, simple, widely available, biologically accurate, composite measure of brain reserve and ability to compensate for ischemic injury, which are essential components underlying the assessment for prestroke morbidity.

To summarize, futile recanalization is common, costly, and complicated. Using historical data to determine prestroke morbidity is haphazard. Recent advances in understanding how imaging brain atrophy and leukoaraiosis may serve as a composite, biologically-based surrogate for prestroke morbidity. Future work to validate and improve the accessibility of using imaging to determine prestroke morbidity may help reduce the currently unsustainable rates of futile transfers and futile recanalization.