the need for fast, precise STROKE MANAGEMENT

marta g. zanchi
nina capital
Published in
11 min readAug 7, 2020

and why we invested in METHINKS

AUGUST 2020

by Marta Gaia Zanchi, Sebastian Anastassiou

Stroke fundamentals

Stroke is one of the leading causes of death globally and the leading cause of chronic adult disability. It is estimated that the health economic cost of stroke amounts to over €45 billion in Europe and $70 billion in the US. A stroke occurs when blood supply to part of the brain is either interrupted or significantly reduced, preventing essential oxygen and nutrients from reaching brain tissue resulting in rapid brain cell death. This is a life threatening medical emergency that requires prompt action and treatment to reduce the risk of permanent brain damage, long term disability and death.

Pathophysiology

Stroke can be classified into three main categories: i) ischemic stroke, caused by interruption of blood supply to the brain, ii) transient ischemic attack (TIA), similar causes as ischemic stroke only short term (“mini stroke”), and iii) haemorrhagic stroke, caused by ruptured blood vessels or abnormal vascular structure.

Ischemic stroke

Ischemic stroke occurs as a result of blockage to cerebral blood flow, the main causes of which are i) thrombosis — arterial plaque build up leading to vessel occlusion, ii) embolism — a blood clot is dislodged from an artery somewhere else in the body and moves to the cerebral arteries where is gets stuck and causes vessel occlusion (can sometimes be other substances, e.g. air, fatty streaks), iii) cerebral hypoperfusion — systemic reduction in blood flow to the brain mostly due to heart failure (cardiac arrest, shock, etc.), and iv) cerebral venous sinus thrombosis — thrombosis of a cerebral vein. When the cause of an ischemic stroke is not readily apparent, it is sometimes termed cryptogenic.

The most common classification system for ischemic stroke is the Bamford system, also known as the Oxford classification system. It classifies stroke on the basis of initial clinically present symptoms (not reliant on imaging) and indicates the extent (severity) of the stroke, affected cerebral area, underlying cause and prognosis. There are four categories, namely i) total anterior circulation infarct (TACI), ii) partial anterior circulation infarct (PACI), iii) lacunar infarct (LACI), and iv) posterior circulation infarct (POCI); the major classifications relate to either large vessel occlusion (LVO) where the blockage is in a major cerebral artery, or whether the blockage is in a smaller cerebral vessel.

In cases of ischemic stroke, the affected area can be further characterised into the core and the penumbra. The core is the irreversibly damaged centre of the affected area and is not salvageable through treatment. The penumbra is the area surrounding the ischemic core that has been damaged but can be salvageable and reversed by rapid treatment to achieve reperfusion. The more time passes, the larger the core grows and the smaller the penumbra area gets, hence the importance of time in stroke treatment — Time is Brain.

Transient ischemic attack (TIA)

A TIA, often called a “mini-stroke”, is a form of ischemic stroke where the blood flow interruption is temporary and short, usually only around five minutes. The American Heart Association defines TIA as “a brief episode of neurological dysfunction with a vascular cause, with clinical symptoms typically lasting less than one hour, and without evidence of infarction on imaging”. TIAs do not cause permanent brain damage but are still classified as a medical emergency as they are indistinguishable from major stroke at initial clinical presentation . Furthermore, they are a strong indicator for future risk of major stroke and should be seriously considered as an early warning sign in need of management.

Haemorrhagic stroke

Haemorrhagic stroke is characterized by bleeding in the brain. There are two main categories, i) intracerebral haemorrhage — most often caused by high blood pressure and aneurysms resulting in leaky or burst arteries, flooding the surrounding brain tissue with blood causing cellular damage, and ii) subarachnoid haemorrhage — less common, also caused by leaky or burst blood vessels resulting in bleeding within the space between the skull and the brain (subarachnoid space).

Clinical presentation

The classic clinical symptoms of stroke involve sudden-onset neurological dysfunction characterised by one or more of i) sudden numbness and/or weakness in the face (facial droop), arm (arm drift) or leg, often contained to one side of the body, ii) sudden confusion, loss of vision, slurred speech or comprehension difficulties, and iii) sudden loss of balance, dizziness, gait abnormalities and coordination difficulties. Haemorrhagic stroke is often further characterised by a sudden severe headache known as a thunderclap headache5,7,.

Epidemiology

As mentioned earlier, stroke is one of the leading causes of death globally and the leading cause of chronic adult disability. In the US, the incidence of stroke has been recorded at approximately 800,000 per year, of which approximately 87% are ischemic and 13% are haemorrhagic. Global stroke incidence is estimated at almost 14 million cases per year with higher haemorrhagic stroke prevalence compared to the US and Europe, thought to be due to apparent higher risk of haemorrhagic stroke seen in South Asian populations. Incidence of stroke is strongly correlated with age, approximately doubling for each decade after the age of 55; around 50% of global stroke cases are within the ages of 45–70 and 40% above the age of 70.11

The main risk factors involved in stroke are age, gender (up to 30% higher risk for males compared to females), ethnicity (higher risk for hispanics and african ethnicities as compared to caucasian), hereditary factors (genetic predisposition), hypertension, and finally lifestyle factors (dietary habits, smoking, alcohol consumption, physical activity). There are also certain diseases that increase risk of stroke such as diabetes, as well as the distinction between acute (first time) and secondary (repeated) stroke — having had a stroke greatly increases the risk of a second stroke incident. ,

Clinical pathway: from diagnosis to treatment

Although clinical guidelines vary across regions and countries, the overall clinical pathway for stroke management is relatively standard. On clinical presentation of suspected stroke (sudden onset neurological symptoms described above), the accepted validated screening tool is known as F.A.S.T.:

  1. Face — is there unilateral facial droop?
  2. Arm — when asked to raise both arms, does one drift downwards?
  3. Speech — is there slurred speech?
  4. Test/Timing — act quickly and call emergency services

Timing is paramount in stroke cases — every second counts as the longer the brain remains without blood supply, the greater severity and risk of permanent brain damage as well as death; patients who receive treatment within three hours of symptom onset have significantly higher chances of recovery with minimal disabilities and every 30 mins past the 3 hour mark reduces the probability of a good clinical outcome by 15%.

Once in the health emergency department, more detailed tools are used which vary across geographies — examples include the ROSIER test (UK) or the NIH Stroke Score test (Global) — all of which use medical history and physical examination to diagnose suspected stroke or TIA. For suspected TIA, brain imaging is not recommended and patients are referred to a specialist and prescribed aspirin. Patients with suspected acute stroke are referred to specialist stroke units and recommended immediate brain imaging using NCCT. NCCT scans enable detection of haemorrhagic stroke with relatively high accuracy so once haemorrhagic stroke is diagnosed, patients undergo immediate treatment to stop the bleeding either through drugs, endovascular procedures or surgery. Once haemorrhagic stroke is ruled out and ischemic stroke is suspected, several other imaging techniques are used for follow-on assessment to further identify the type and location of the ischemic stroke (see Bamford classification above) in order to inform the most appropriate treatment. Depending on the classification of the ischemic stroke as diagnosed through additional imaging, the treatments are often one or a combination of i) aspirin, ii) thrombolysis using tissue plasminogen activator (t-PA) drug such as Alteplase to break down the occlusion material (recommended up to 4.5hrs of symptom onset), and iii) thrombectomy (endovascular treatment, EVT) — mechanical removal of the clot if the occlusion is located in a major cerebral artery (i.e. an LVO), recommended within 6 hours of symptom onset and performed in what is commonly known as the angiography suite (“angio suite”).,

Although most guidelines suggest performing contrast CT (depending on time passed since symptom onset) in the form of CT Angiography (CTA) or CT Perfusion (CTP), there are several other imaging techniques that are also used, mainly based on MRI.

Contrast Computed Tomography

Contrast CT involves CT imaging using intravenous injected contrast agents for enhanced imaging. The two main types of contrast CT used in stroke diagnosis are i) CT Angiography (CTA) — enables the visualisation of the cerebral vasculature and clear identification of vessel occlusions, and ii) CT Perfusion (CTP) — enables the visualisation of cerebral blood flow to identify areas of low blood perfusion and the penumbra, and is particularly useful for difficult-to-diagnose small vessel occlusions.

Although many hospitals have contrast CT capabilities, there are many that do not, particularly the smaller hospitals and stroke centres. The main reasons for this are that i) trained technicians are required, ii) radiologists are needed to monitor the contrast agent administration (allergic reactions are common), and iii) specialists are needed to interpret the contrast CT images. From a patient standpoint, the main limitations are allergic reactions, patients with underlying renal inefficiencies and possible high radiation doses (particularly with CTP).

Magnetic Resonance Imaging

Various MR based imaging techniques are used when available in stroke diagnosis. Traditional MRI, MR Diffusion and MR Angiography are powerful tools to differentially diagnose more complex and difficult to diagnose stroke cases. However, the major limitations of MR based imaging for stroke management are i) lack of availability — few emergency stroke centres have an MRI machines (with the exception of France), and ii) time component — although the actual imaging time is short, the pre-workup is too long as patient histories are needed to ensure no pacemakers and metallic implants are present, in which case the patient is ineligible for MRI.

Other imaging modalities include sonography and catheter-based angiography that are used in certain rarer cases where other modalities are not possible or diagnosis is less clear cut and more information is needed.

The main takeaway is that CTA and CTP are the gold standard imaging techniques for stroke diagnosis after NCCT.

Clinical pathway: misdiagnosis in stroke care

As haemorrhagic stroke is relatively well diagnosed, the majority of misdiagnosis revolves around ischemic stroke. Failure or delay to diagnose ischemic stroke (i.e. false negatives known as “stroke chameleons”, SC) prevents administration of time-sensitive treatments which results in adverse clinical outcomes including increased risk of stroke recurrence and increased mortality rates compared to stroke patients who are accurately diagnosed. Conversely, false positive diagnosis or ‘overdiagnosis’ of ischemic stroke (known as “stroke mimics”, SM), has very low negative impact on patient outcomes as administering thrombolytics to SM cases do not result in many cases of adverse effects. Furthermore, the EVT procedure is designed such that LVO confirmatory diagnosis is done prior to entering the cerebral vasculature. This, in combination with the life saving benefits of early treatment of stroke patients, has led to clinical guidelines advocating treatment even in possible SM cases rather than risk delaying treatment.

Stroke has been reported as the fourth most misdiagnosed disease type, with misdiagnosis rates varying greatly between studies; SM misdiagnosis rates have been reported at 30% — 43%, with SC misdiagnosis rates at 2% — 26%. Although not entirely clear, it is believed that the majority of misdiagnosis is attributable to a combination of system-related (technical failure, organizational errors), and cognitive factors (flaws in data gathering, knowledge, and synthesis). The key takeaway message here is the overall high level of misdiagnosis that has important implications for patient survival, patient care, as well as cost to healthcare systems.

Clinical pathway: the hub and spoke model in stroke care

The “hub and spoke” model is a very well used system to organise healthcare services in a manner that ensures high quality care delivery in the most efficient way possible. The basic design involves a centralised anchor centre (the “hub”) that offers a full array of services, complemented by several satellite centres (the “spokes”) that offer a more limited array of services and route patients requiring more intensive and complex treatment/services to the hub.

In the case of stroke care, the hub and spoke model is arrange by a central large comprehensive stroke centre (the hub), which provides advanced diagnostics, treatments and highly trained personnel for stroke management, and is connected to several primary stroke centres (the spokes), which provide initial acute diagnostics and care for stroke patients.

The typical journey of a patient would involve i) emergency services as first responders making initial assessments (FAST) on the way to the closest primary stroke centre (spoke) ii) at the primary stroke centre, NCCT will be performed alongside assessment scales (ROSIER, NIHSS) to triage patients into either ischemic, haemorrhagic, inconclusive or non-stroke groups, iii) the ischemic suspected/ inconclusive patients will then get transferred to the comprehensive stroke centre (hub) for CTA and CTP imaging for final diagnosis before treatment. This journey describes a ‘typical’ situation and there are several nuances that in reality may result in slightly different scenarios, for example, if the comprehensive stroke centre is closest, the patient will go directly there instead of through a primary stroke centre.

Although the hub and spoke model in stroke care is working well, there are limitations related to effective communication between the primary and comprehensive stroke centres to maintain efficient and consistent care to patients being transferred from one centre to the other. Furthermore, given the importance of time in stroke care, unnecessary transfers due to misdiagnosis and long transfer times are a major pain point for stroke centre networks.

Stroke treatment (thrombectomy) nearly always improves outcomes. Getting the patient to the operating room in the most expeditious way possible is key. Solutions are needed to speed up the workflow, triage patients in situ etc... and the same logic can be applied to other medical emergencies that are very time dependent (e.g. cardiac arrest).

Trends in stroke

The burden of stroke is measured by Disability-Adjusted Life Years (DALYs), which is a measure of overall disease burden accounting for the number of years lost due to early death, and the number of years lived with disability. Although there are decreasing trends in stroke mortality rates and mortality-to-incidence ratios, the global burden of stroke as reported by the absolute number of people affected every year, stroke survivors, related deaths, and DALYs lost are significant and trending upwards. Forecasting these trends in stroke incidence, mortality, and DALYs through to 2030, there will be almost 12 million stroke deaths, 70 million stroke survivors, and more than 200 million DALYs lost globally. Furthermore, these trends are augmented in low to medium income countries, likely due to disparities in health services and stroke management resources.

There is also a growing trend of evidence and backing in the stroke community for the direct transfer of LVO patients to angiography suites for EVT without the need for advanced imaging. Several clinical trials are being run by research groups as well as companies such as Phillips and Siemens to validate the effectiveness of this workflow change,,. This complements the growing trends in the use and adoption of AI in stroke care, which have a direct relationship with triaging LVO patients directly to treatment. The National Institute of Clinical Excellence (NICE) in the UK recently released an interesting briefing on the matter.

Concluding remarks

In summary, the stroke management ecosystem presents several challenges and unmet needs that, if addressed, would significantly benefit the entire landscape, the most compelling of which relate to:

  1. Timely stroke diagnosis — time is the most critical factor in stroke management; the first three hours from symptom onset are the most critical, with a target treatment timeline of no more than 4.5–6 hours post symptom onset. This means effective triaging to minimise the number of diagnostic tests and start life saving treatment as early as possible is the major need for stroke management.
  2. Misdiagnosis — having one of the highest rates of misdiagnosis, stroke management is in need of tools to reduce misdiagnosis and improve positive clinical outcomes for patients.
  3. Operational efficiency — given the crucial nature of time in stroke management, maximising operational efficiency is paramount, both in terms of increasing diagnostic operational capacities of primary stroke centres, as well as improving the integrity of patient data flow through the hub and spoke networks of stroke management enabling better treatment coordination.

ENTER: METHINKS

a way to deliver a differential diagnosis and inform clinical decisions in stroke management in order to reduce the time to personalized, precise treatment.

www.methinks.ai​​

--

--

nina capital
nina capital

Published in nina capital

nina capital is a new venture capital firm investing at the intersection of healthcare and deep technology.

marta g. zanchi
marta g. zanchi

Written by marta g. zanchi

health∩tech. recognizing the need = primary condition for innovation. founder, managing partner @ninacapital

No responses yet