After the statistical analyses were run on the subcortical segmentation data of the ventricular system some interesting trends were found. Among right hemisphere lateral ventricle data, a trend was seen in the enlargement of the lateral ventricles in the TBI group as compared to controls. This trend came extremely close to significance (p=.063) suggesting a possible connection between traumatic brain injury in the developing brain and the enlargement in volume of the lateral ventricles, combined with a possible retreat in subcortical grey matter. Interestingly, the left hemisphere ventricular volumes did not approach significance as did the right hemisphere. Due to the small number of subjects and a relatively high amount of variance found in the data received, a Levene's Test for Equality of Variances was run in order to create a more accurate p-value while accounting for the variance seen. The Levene's Test showed that variance did impact the initial data but after the recalculations were run, the trend of enlargement was still seen, though slightly less significantly in the statistics. 
    In addition to a trend in the enlargement of the right hemisphere lateral ventricles, a similar trend was seen in the enlargement of third ventricle volume among the TBI group. The same Levene's Test was run to ensure accuracy and though not reaching levels of significance, the p-values between groups still showed a possible trend of volume enlargement of the third ventricle (p=.096). For unknown reasons, the trend in third ventricle volume expansion between the TBI and control groups did not relate to the third ventricle length measurements that were found through hand measurements on Mango. Individual volumes of third ventricles were also not found to be related to ventricular length as measured by Mango. Due to the lack of correlation between ventricular length and overall volume in both a group and individual measurements, it is doubtful that length measurements are an effective method of determining ventricular volume. It is possible that the volume enlargement occurred at a different level of the brain than that measured between the anterior and posterior commissures. It is also reasonable that the changes occur at all levels of the third ventricle and that the expansion is too minimal to be measured lengthwise but can be seen in overall volume. Such findings suggest further research in order to better understand the where the possible areas of gray matter retreat could best be found and measured. 

Discussion

    The trend data discovered in this study leaves room for some possible conjectures about neurostructural processes following pediatric TBI. Though no data was found to be significant, the trends indicate a possible volumetric expansion seen in both right hemisphere lateral ventricles and the third ventricle in the pediatric populations. Such trends are both concerning and intriguing. Concern for the long-term welfare of pediatric patients who may be suffering from a retreat in subcortical gray matter is a prominent issue. This being especially true due to the importance of structures surrounding the expanding ventricles, especially in the area of the third ventricle. If such gray matter destruction were found, it would most likely involve the hypothalamus, a structure key to a plethora of important functions key to life including homeostasis of temperature and appetite along with being the main connection between the nervous and endocrine systems. Early damage to the hypothalamus could lead to lifelong challenges in TBI patients. Thus, more research is crucial to better define whether such volumetric expansion is actually occurring, whether it is in fact due to subcortical gray matter lesioning and destruction, and which areas of gray matter are at risk.
    The author believes that a variety of studies are needed to further understanding in this area. Such studies may begin with a larger sample size of TBI and control patients so as to give greater power to statistical analyses and a clearer view of the data. Additionally, it may be prudent to observe neurostructural changes in a longitudinal fashion in order to track such changes over time and to more clearly track areas or motives for such volumetric expansion. It may also be advantageous to examine varied age ranges including pediatrics, young and middle age adults, and older adults in order to determine if such volumetric expansion may differ by age. Interesting sex-related differences have also been documented extensively in both TBI occurrence and severity and neurostructural changes should be examined as part of those differences as well.
    Though limitations of this study did exist, precautions were taken to minimize as many as could be controlled. The small number of participants in this study (n=20) did have an effect on the significance of the data and did give much greater power to the variance seen between individuals. Though all data was examined for accuracy, there has been some debate on the accuracy of subcortical segmentation methods in neuroimaging. Though unlikely to be perfect in their results, repeated studies have found their measurements to be acceptably accurate and it is expected that the results in this study are no different. Though a perfect ventricular volume cannot be measured through current imaging methods, we expect the data to be acceptable and within reason. As with all imaging studies, certain data may be rendered inaccurate through lack of clarity in the images or patient movement in the scanner, but steps have been taken to moderate these inaccuracies. To the author's knowledge, despite the limitations, the data was secured and calculated in an acceptable and useful manner.
References