Mechanical thrombectomy: Determining the proportion of eligible acute ischemic stroke patients in the cohort of single academic stroke center

Estimated number of eligible patients for mechanical thrombectomy based on NIHSS and population-based Brest stroke registry

BACKGROUND

Since 2015, mechanical thrombectomy (MT) is indicated as a treatment for patients with large vessel occlusion (LVO) at the acute phase of ischemic stroke. However, the number of stroke patients eligible for MT is poorly known.

OBJECTIVE

The objective of our study was to estimate the number of patients eligible for thrombectomy within the first 24hours of an ischemic stroke, based on the clinical National Institute of Health Stroke Scale (NIHSS).

METHOD

Our study concerned all ischemic strokes which occurred between January 2013 and December 2016 recorded in the population-based Brest Stroke Registry (BSR). Based on positive predictive value and negative predictive value from articles evaluating the performance of a defined NIHSS threshold to identify LVO, we first estimated the frequency of patients with LVO and then the frequency of patients eligible for MT depending on pre-stroke modified Rankin score (mRS). Our results were extrapolated to regions of metropolitan France. Two scenarios were considered: one called "stringent criteria" with mRS ≤1 and one called "real-life" criteria with mRS ≤2.

RESULT

We analyzed data from 2,025 ischemic strokes with symptom onset ≤24hours. No statistical difference between patient characteristics according to the time of hospital admission (≤6H vs. 6-24H) was observed. Based on NIHSS scores, between 23.90% and 44.20% of ischemic strokes admitted within the first six hours had LVO clinical characteristics. Among them, 14.53% to 26.87% met the ``stringent eligibility'' criteria for MT and 16.9 to 31.25% for ``real-life'' criteria. Eligible patients represented 6.32% to 11.70% of all ischemic strokes, irrespective of admission time. In France, 75 to 162 persons per million inhabitants per year were eligible for endovascular therapy, depending on including criteria. Based on activity levels recorded by the French Neuroradiology Society (SFNR) in 2018, the estimated needed increase in MT showed a heterogeneous pattern region-by-region, with the greatest need in Brittany, Pays de la Loire, and Corsica.

CONCLUSION

Based on NIHSS, our study provides coherent information concerning the estimated number of MT procedures to be performed in France: 4,877 to 10,494 ischemic strokes would be eligible each year in metropolitan France compared to the 6,596 thrombectomy procedures actually performed in 2018. Depending on the region, an estimated 10-20% to 90-100% increase in MT activity would be necessary to meet patient needs. These data suggest that there is still room for improvement in thrombectomy activity, particularly in certain regions of France, to allow equal access to MT to the entire French population.

Stroke patients from rural areas have lower chances for long-term good clinical outcome after mechanical thrombectomy

BACKGROUND

This study evaluated 3-months clinical outcome after mechanical thrombectomy (MT) in stroke patients transferred to a comprehensive stroke center (CSC) from a rural and urban areas in a Lubelskie province, the third largest province in Poland.

MATERIALS AND METHODS

Acute stroke patients with a premorbid modified Rankin scale (mRS) score 0-2 who were admitted within 6 h after stroke onset and treated with MT between 2016 and 2020 were retrospectively analyzed. Patients from rural and urban areas transported directly to CSC were compared regarding the onset-to-groin time, reperfusion rate, symptomatic intracranial hemorrhage (sICH) and favourable clinical outcome (modified Rankin Scale score 0-2) 3-months after MT.

RESULTS

A total of 398 patients were analyzed: 179 from rural areas (RA) and 219 from urban areas (UA). There was no significant difference in baseline neurological deficit expressed in The National Institutes of Health Stroke Scale (median 18.4 for RA patients versus 18.1 for UA patients, p = 0.70). Time from stroke onset to groin puncture was significantly shorter in the UA patients (median 197.3 min versus 219.6 min, p = 0.004). There was a significant difference in 3 months favourable clinical outcome between these two groups (31.3% of RA patients versus 42.5% of UA patients, p = 0.021) and full recovery rates (5.6% of RA patients versus 15.0% of UA patients, p = 0.002). The rate of sICH and 3-months mortality was similar in both groups (7.3% of RA patients versus 8.7% of UA patients, p = 0.61% and 21.8% of RA group vs. 22.4% of UA group, p = 0.88, respectively).

CONCLUSION

Stroke patients from RA undergoing thrombectomy had worse functional outcome compared to UA patients. Since the benefit of MT is time dependent, urban-rural differences in stroke outcome probably result from the longer time from stroke onset to reperfusion treatment in RA patients.

"Strokenomics": bending the cost curve in stroke care

OBJECTIVE

The mortality rates for stroke are decreasing, yet it remains a leading cause of disability and the principal neurological diagnosis in patients discharged to nursing homes. The societal and economic burdens of stroke are substantial, with the total annual health care costs of stroke expected to reach $240.7 billion by 2030. Mechanical thrombectomy has been shown to improve functional outcomes compared to medical therapy alone. Despite an incremental cost of $10,840 compared to medical therapy, the improvement in functional outcomes and decreased disability have contributed to the cost-effectiveness of the procedure. In this study the authors describe a physician-led device bundle purchase program implemented for the delivery of stroke care.

METHODS

The authors retrospectively reviewed the clinical and radiographic data and device-associated charges of 45 consecutive patients in whom a virtual "stroke bundle" model was used to purchase mechanical thrombectomy devices.

RESULTS

Use of the stroke bundle to purchase mechanical thrombectomy devices resulted in an average savings per case of $2900.93. Compared to the traditional model of charging for devices à la carte, this represented an average savings of 25.2% per case. The total amount of savings for these initial 45 cases was $130,542.00. Thrombolysis in Cerebral Infarction scale grade 2b or 3 recanalization occurred in 38 patients (84.4%) using these devices.

CONCLUSIONS

Purchasing devices through a bundled model resulted in substantial cost savings while maintaining the therapeutic efficacy of the procedure, further pushing the already beneficial long-term cost-benefit curve in favor of thrombectomy.

Quantifying the Impact of Chronic Ischemic Injury on Clinical Outcomes in Acute Stroke With Machine Learning

Acute stroke is often superimposed on chronic damage from previous cerebrovascular events. This background will inevitably modulate the impact of acute injury on clinical outcomes to an extent that will depend on the precise anatomical pattern of damage. Previous attempts to quantify such modulation have employed only reductive models that ignore anatomical detail. The combination of automated image processing, large-scale data, and machine learning now enables us to quantify the impact of this with high-dimensional multivariate models sensitive to individual variations in the detailed anatomical pattern. We introduce and validate a new automated chronic lesion segmentation routine for use with non-contrast CT brain scans, combining non-parametric outlier-detection score, Zeta, with an unsupervised 3-dimensional maximum-flow, minimum-cut algorithm. The routine was then applied to a dataset of 1,704 stroke patient scans, obtained at their presentation to a hyper-acute stroke unit (St George's Hospital, London, UK), and used to train a support vector machine (SVM) model to predict between low (0-2) and high (3-6) pre-admission and discharge modified Rankin Scale (mRS) scores, quantifying performance by the area under the receiver operating curve (AUROC). In this single center retrospective observational study, our SVM models were able to differentiate between low (0-2) and high (3-6) pre-admission and discharge mRS scores with an AUROC of 0.77 (95% confidence interval of 0.74-0.79), and 0.76 (0.74-0.78), respectively. The chronic lesion segmentation routine achieved a mean (standard deviation) sensitivity, specificity and Dice similarity coefficient of 0.746 (0.069), 0.999 (0.001), and 0.717 (0.091), respectively. We have demonstrated that machine learning models capable of capturing the high-dimensional features of chronic injuries are able to stratify patients-at the time of presentation-by pre-admission and discharge mRS scores. Our fully automated chronic stroke lesion segmentation routine simplifies this process, and utilizes routinely collected CT head scans, thereby facilitating future large-scale studies to develop supportive clinical decision tools.