Synopsis

Keywords: Stroke, Treatment

Cerebral edema occurs after stroke. Mannitol is commonly used dehydrating agent for effectively improving cerebral edema. However, the edema state after mannitol administration and its effectiveness at the post-acute stages remain unclear. Recently, we have demonstrated free-water elimination and mapping (FWET₂) can be used to characterize vasogenic edema in stroke animal models. The aim of this study was to quantitatively evaluate mannitol-induced effects on ischemic rat brains at different time points by using FWET₂. We attempted to examine if free water fraction change could reflect mannitol-induced effects on infraction and correlate free water fraction change with total infarct volume.

Introduction

Cerebral edema typically occurs after acute stroke, which is defined as an excess fluid accumulation within brain parenchyma^.1,2 Mmannitol, hyperosmolar agents, is commonly used as treatment to reduce brain edema.^3,4 However, the edema state after mannitol administration and its effectiveness administrated in the post-acute period remain unclear. Recently, we have demonstrated FWET₂-provided free water fraction (f_w) associates with the pathological changes linked to vasogenic edema evolution in ischemic rat brain.⁵ Therefore, this aim of the study was to evaluate the mannitol-induced effects on brain infarction at acute and post-acute stages of stroke rats using FWET₂. We attempted to examine if free water fraction change (Δf_w) could reflect mannitol-induced effects on ischemic infraction and correlate the free water fraction change with total infarct volume.

Methods

Twelve adult SD rats (300-450 g) underwent 90 minutes of right MCA occlusion.⁶ Rat was injected with a single dose of 20 % mannitol solution⁸ through intravenous injection. For each experiment, two MRI scans were performed, before and 20 minutes after mannitol (MTL) injection. MRI were performed on day 1 (n=5), day 5 (n=6), and day 26 (n=6) after stroke on an in-house 3T MRI system.⁷ An ultra-high-strength gradient coil with a maximum strength of 675 mT/m (Resonance Research Inc., MA, USA) and a custom transmit/receive surface coil were utilized. Diffusion-weighted images were acquired using a Stejskal-Tanner EPI pulse sequence implemented with the b-value (ms/μm²)/number of diffusion directions: 0/8, 0.5/12, 1.0/26, and 2.0/40 (b = 2.0 ms/μm² was not used in this work). The acquisition parameters were TR=9 s; TE=50, 100 ms; Δ=24 ms; δ= 3 ms; one repetition. Addition eight b=0 images (eight repetitions) with TE=70, 90, 110, 130 ms were collected. Other parameters were: FOV = 25× 25 mm²; matrix-size = 96 ×96; voxel-size = 0.26 × 0.26 ×1 mm³; 20 image slices. T₂-weighted images were also acquired. FWET₂-DTI model analysis was performed as previously described.⁵ All datasets were first denoised, corrected for EPI, eddy-current distortions, and with linear and non-linear registrations by ANTs software.⁶ Maps of fraction anisotropy (FA), mean diffusivity (MD), T₂ and f_w were reconstructed. The difference map of free water fraction (Δf_w) between after and before MTL was calculated. Regions-of-interest (ROIs) of whole cortical infarction were determined considering both FA, MD, and T2 maps before MTL. Total infarction volume was multiplication of total infarct area with slice thickness. Whole infarction f_w and ∆f_w histograms were analyzed. The free water fraction change (∆f_w) were evaluated. Correlation analysis between ∆f_w and total infarction volume was performed. Two-tailed paired t-test was performed for statistics.

Results

Figure 1 shows representative f_w maps before and after MTL, and ∆f_w maps at different time points. On day 1, f_w was increased particularly in core lesion followed by dehydrated after MTL. On day 5, f_w was heterogeneously increased in lesion followed by a rapid rebound in cerebral edema after MTL. Finally, on day 26, f_w was great in lesion followed by an minor increase with large variation after MTL. Figure 2 shows quantification assessments of f_w_mean (mean value of f_w histogram) before and after MTL, and ∆f_w_mean (mean value of ∆f_w histogram) at different time points. Before MTL, f_w_mean shows significant differences from day 1 vs. 26 and day 5 vs. 26. After MTL, f_w_mean shows significant differences between either two of three time points (all p<0.005, Fig. 2A). In addition, ∆f_w_mean was calculated to examine the mannitol-induced effects on infarction. With time course, ∆f_w_mean was -0.013±0.009, 0.014±0.012, and 0.010±0.035 for day 1, 5, and 26, respectively (Fig. 2B). ∆f_w_mean shows significant difference from day 1 vs. 5 (p<0.005, Fig. 2B). Table 1 listed the comparison of other variables from ∆f_w histograms at different time points. Note that standard deviation (SD) of ∆f_w histograms (i.e., ∆f_w_SD) was around twice larger on day 26, as compared with that on day 1 and 5. Figure 3 shows the association between ∆f_w and total infarction volume. The Pearson correlation coefficients (r) were 0.23, -0.64, and 0.39 for day 1, 5, and 26, respectively.

Discussion and Conclusions

We have demonstrated that FWET₂ revealed spatialtemporal characterization of free water fraction in ischemic infarction before mannitol injection at different stages, consistent with our previously published paper.⁵ Evaluation of ∆fw_mean revealed that dehydration at acute and a rapid rebound of cerebral edema at subacute, while evaluation of ∆f_w_SD revealed large variation at chronic. It is known that mannitol has action effect (dehydration) and adverse effect (rebound water due to mannitol accumulation). In addition, blood-brain barrier (BBB) pathology and its permeability was developed and varied throughout the different stages of ischemic stroke.⁹ Our results suggest the free water fraction change may reflect mannitol-induced effect on cortical infarction at different stages, which were associated with BBB disruption and its permeability. Interestingly, the mannitol-induced rehydration within ischemic lesion correlates with total infarct volume at subacute stage. It may relate to the development of tissue pathological heterogeneity of lesion. In summary, importantly, FWET2-provided free water fraction is reliable for evaluating and reflecting mannitol-induced effects on cerebral infarction in stroke rats. This work may promote the management and improvement of treatment strategies.

Acknowledgements

No acknowledgement found.