Reliability of infarct volumetry: Its relevance and the improvement by a software-assisted approach

Abstract

Despite the efficacy of neuroprotective approaches in animal models of stroke, their translation has so far failed from bench to bedside. One reason is presumed to be a low quality of preclinical study design, leading to bias and a low a priori power. In this study, we propose that the key read-out of experimental stroke studies, the volume of the ischemic damage as commonly measured by free-handed planimetry of TTC-stained brain sections, is subject to an unrecognized low inter-rater and test-retest reliability with strong implications for statistical power and bias. As an alternative approach, we suggest a simple, open-source, software-assisted method, taking advantage of automatic-thresholding techniques. The validity and the improvement of reliability by an automated method to tMCAO infarct volumetry are demonstrated. In addition, we show the probable consequences of increased reliability for precision, p-values, effect inflation, and power calculation, exemplified by a systematic analysis of experimental stroke studies published in the year 2015. Our study reveals an underappreciated quality problem in translational stroke research and suggests that software-assisted infarct volumetry might help to improve reproducibility and therefore the robustness of bench to bedside translation.

Beta adrenoceptor blockade ameliorates impaired glucose tolerance and alterations of the cerebral ceramide metabolism in an experimental model of ischemic stroke

“…Volumetry of the ischemic lesion, the contra- and the ipsilesional hemisphere was performed using a previously established automated ImageJ macro (ImageJ software, National Institutes of Health, Bethesda, USA). 27 Lesion size was corrected for edema by multiplying the infarct volume by the ratio of the contralateral to the ipsilateral hemisphere volume.…”

Gene Expression Dynamics at the Neurovascular Unit During Early Regeneration After Cerebral Ischemia/Reperfusion Injury in Mice

“…Individual stroke volumes could not be quantified during the isolation process without risking loss of vital material and artificial tissue reactions. Medium stroke sizes in our surgery setup covering about 60% of the hemisphere have been reported before (Friedländer et al, 2017). To estimate stroke volumes in the present study, male C57B/6J mice were operated by RK under same conditions as the mice used for mRNA sequencing.…”

“…Subsequently, brain slices were stained in 2% 2,3,5-Triphenyl-tetrazolium chloride (TTC) at 37 • C for 5 min and imaged on a flatbed color scanner (CanoScan LiDE 100; Canon, Tokyo, Japan). The hereby obtained Photoshop images were used for semi-automatical stroke volume calculation by an ImageJ macro established by Friedländer et al (2017). Individual results are summarized in Supplementary Figure IV (mean stroke volume 57.2%, median 55.6%).…”

Image-based stroke rat brain atrophy volume and infarct volume computation

“…In addition to the selection of animal model, how to appropriately interpret the animal experimental results is another important issue. Some studies, for example, employed well-trained animal physiologist to determine the area of rat ischemic stroke, and some employed public image procession program such as ImageJ to determine its severity [5,13,14,34]. In the preclinical animal studies of stroke drug development, there are massive demands for stroke image analysis.…”

“…In the preclinical animal studies of stroke drug development, there are massive demands for stroke image analysis. However, few attempts have been made to develop automatic calculation algorithms to deal with this problem [13,23,24]. Therefore, it is critical to develop methods that can assess the effects of drugs or treatments on stroke animal models.…”