Yu XIE^{1}, Catherine Oppenheim^{2}, Francis Guillemin^{3}, Vincent Gautheron^{2}, Benjamin Gory^{4}, Hélène Raoult^{5}, Sébastien Soize^{6}, Bailiang CHEN^{7}, Jacques Felblinger^{1,7}, Gabriela Hossu^{7}, and Serge Bracard^{1,8}

The association between pretreatment ischemic volume (PIV) measured on diffusion weighted images (DWI) and functional outcome after mechanical thrombectomy is of great clinical importance but has yet to be determined. We analyzed 298 ischemic stroke patients from the multicentric study THRACE. Our results showed that increased PIV was an independent predictor for a lower probability of functional independence, a less favorable degree of disability, and a higher mortality rate. PIV measured on DWI is a valuable early predictor for functional outcome in ischemic stroke patients, and thus can contribute to patient selection for optimal therapeutic intervention.

**Introduction**

**Methods**

*Patients:* 298 anterior circulation ischemic
stroke patients who had pretreatment DWI were included in this work. They were
all recruited in a multicentric study THRACE^{3}, which was a
randomized controlled trial done in 26 centers in France, comparing intravenous
thrombolysis plus mechanical thrombectomy (IVTMT) and intravenous thrombolysis
alone (IVT) in acute ischemic stroke patients. It provides a unique opportunity
to investigate the DWI-PIV, because of the use of DWI in pretreatment imaging
in a large proportion of patients and the absence of exclusion criteria on the
basis of PIV.

*Data analysis*: Pretreatment lesion regions
were semi-automatically segmented after applying a threshold of apparent
diffusion coefficient (ADC) of 0.6×10^{-3 }mm^{2}/s on ADC maps,
which were reconstructed from DWI acquisitions. DWI-PIVs were automatically measured
based on the delineated lesion regions. All the analyses were performed using
Olea Sphere^{®} (Olea Medical SAS, La Ciotat, France).

*Outcome assessment and statistical analysis*: Primary
outcome was the proportion of
patients achieving functional independence at 3 months, defined by the modified
Rankin Scale (mRS) 0-2. Secondary outcomes were (1) degree of disability at 3
months, assessed by the ordinal mRS from 0 to 6 with ranks 5 and 6 combined into
a single category; and (2) mortality at 3 months. Logistic regression was
applied to model clinical outcome as a function of DWI-PIV. α=0.05 was chosen as the significance
level, statistical analyses were performed using statistical software R (version
3.3.2)

**Results**

Of 298 patients, with median ischemic volume 17.2 mL (interquartile range [IQR] 9.2-51.8) and median mRS 2 (IQR 1-4), 51.0% achieved functional independence.

The distribution of DWI-PIV for each mRS level is shown in Figure 1. Increased ordinal mRS was associated with a larger DWI-PIV (Spearman’s ρ correlation coefficient 0.34; p<0.0001). Increased DWI-PIV was an independent predictor for a lower probability of functional independence (odds ratio [OR] 0.90 [95% confidence interval [CI] 0.81-0.99] per 10 mL, p<0.001), a less favorable degree of disability (common OR [cOR] 0.86 [95% CI 0.81-0.90] per 10 mL, p<0.001), and a higher mortality rate (OR 1.21 [95% CI 1.08-1.37] per 10 mL, p<0.001). The associations between DWI-PIV, age, baseline NIHSS score, and adjusted probability of functional independence were depicted by three-dimensional scatterplots with fitted surfaces (Fig 2). No significant treatment-by-dichotomized DWI-PIV (dichotomized by PIV of 70 mL and 100 mL) interaction for functional independence was observed.

With the increase of DWI-PIV, the probability of achieving functional independence decreased in both the IVTMT group and the IVT group (Fig 3). For the IVTMT group patients: OR 0.87 (95% CI 0.77-0.98, p=0.02) per 10 mL; for the IVT group patients: OR 0.89 (95% CI 0.80-0.97, p=0.01) per 10 mL; for the IVTMT group patients with substantial reperfusion (defined as a modified Thrombolysis in Cerebral Infarction (mTICI) scale score of 2b or 3): OR 1.26 (95% CI 1.00-1.59, p=0.04) per 10 mL.

**Discussion**

**Conclusion**

We thank Farhat Benbelkacem and Raphaël Petazzoni (Olea Medical) for their help in image postprocessing.

1. Goyal M, Menon BK, van Zwam WH, et al. Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016;387(10029):1723-1731.

2. Edlow, J. A. Evidence-based guideline: the role of diffusion and perfusion MRI for the diagnosis of acute ischemic stroke: report of the Therapeutics and Technology Subcommittee of the American Academy of Neurology. Neurology. 2011;76(23):177-185.

3. Bracard S, Ducrocq X, Mas JL, et al. Mechanical thrombectomy after intravenous alteplase versus alteplase alone after stroke (THRACE): a randomised controlled trial. Lancet Neurol. 2016;15(11):1138-1147.

4. Gilgen MD, Klimek D, Liesirova KT, et al. Younger Stroke Patients With Large Pretreatment Diffusion-Weighted Imaging Lesions May Benefit From Endovascular Treatment. Stroke. 2015;46(9):2510-2516.

Box-and-whisker plot showing the distribution of the pretreatment ischemic
volume for each modified Rankin Scale (mRS) level. The boundaries of the box
indicate the 25th and 75th percentiles and the black dots indicate the median
values. The whiskers represent the lowest and highest values in the 25th
percentile minus 1.5 interquartile range (IQR) and 75th percentile plus 1.5
IQR, respectively. Circles indicate outliers.

Three-dimensional scatterplots with fitted surfaces
illustrating the association between age (A) or baseline NIHSS score (B), and
pretreatment ischemic volume, and adjusted probability of functional
independence (mRS 0-2) at 3 months. The colored crosses indicate the estimated
probability of functional independence for each patient in our cohort.

Probability of functional
independence according to pretreatment ischemic volume, by treatment group.
Curves were obtained from binomial logistic regression of functional
independence at 3 months on pretreatment ischemic volume as a continuous
variable, adjusted for age, baseline NIHSS score, occlusion location of proximal portion of middle cerebral artery, and
blood glucose level. The solid lines showed the model results from the logistic
regression analysis, with 95% CIs shown in the color-corresponding dashed
lines. Red lines indicate patients in IVT group; green lines indicate patients
in IVTMT group; blue line indicates patients in IVTMT group who achieved
substantial reperfusion.