Abstracts

Rationale: For patients with medically intractable focal epilepsy, the best option for achieving seizure control is often surgical resection. In surgical planning, the potential for seizure reduction must be weighed against the risk of cognitive loss. The role that clinical and demographic factors play in predicting cognitive outcome is well established; however, little is known about the role of cross-hemispheric white matter in promoting functional reorganization after surgery. In this study we measured the midsagittal cross-sectional area of the corpus callosum (CC) on pre-surgical MRI to investigate whether this property is related to changes in working memory following surgery.Methods: A pre- and post-surgical neuropsychological test battery was obtained in 15 patients (9 males/6 females) who underwent temporal (n = 9), frontal (n = 4), temporal and frontal (n = 1) or parietal lobe (n = 1) resective surgery at NYU Langone Medical Center. Pre-surgical whole-brain T1-weighted 3D MRIs were acquired on all participants from the same dedicated research scanner. The midsaggital CC cross-sectional area was delineated and measured automatically on the MRI using yuki (www.nitrc.org/projects/art), an automatic CC segmentation algorithm, described by Ardekani et al. 2012 (Figure 1A). The Working Memory Index (WMI) from the Wechsler Adult Intelligence Scale was used to probe change in concentration/working memory abilities (postsurgical WMI presurgical WMI). Positive scores represent post-surgical improvement in working memory and negative scores post-surgical decline. A regression model was used to determine whether presurgical CC area is associated with a pre- to post-surgical change in working memory abilities, after controlling for overall brain volume, age at surgery, presurgical WMI, and duration of epilepsy.Results: Working memory improved in 9 patients (6 temporal, 1 frontal, 1 temporal/frontal and 1 parietal) following surgery and declined in 6 patients (3 temporal and 3 frontal). Pre-surgical CC area predicted pre- to post-surgical change in WMI scores, after controlling for brain volume [B = 0.82, t(14) = 4.32, p = 0.001], and accounted for 56% of the variance in working memory change scores (see Figure 1B). The relationship was positive, indicating that for every 1-unit increase in CC area, there is a 0.14-unit increase in working memory change scores. This relationship remained significant after controlling for age at surgery, presurgical WMI, and seizure duration [B = 0.75, t(14) = 3.56, p < 0.01].Conclusions: Our study demonstrates that the relative size of the corpus callosum is a predictor of postsurgical cognitive recovery. This relationship remained strong after controlling for established predictors such as age, presurgical ability, and duration of epilepsy. Our study indicates that morphometric analysis of the CC using MRI may be useful for predicting greater potential for neuroplasticity and functional reorganization following brain surgery. Ardekani et al, Cereb Cortex (2012) doi: 10.1093/cercor/bhs253