Hajimu Goto MD, Presenter: Nothing to Disclose

Kazunari Ishii MD, Abstract Co-Author: Nothing to Disclose

Hiroaki Kazui MD, Abstract Co-Author: Nothing to Disclose

Masaaki Hashimoto MD, Abstract Co-Author: Nothing to Disclose

Kohei Hayashida MD, PhD, Abstract Co-Author: Nothing to Disclose

Kazuo Hashikawa MD, Abstract Co-Author: Nothing to Disclose

Chia-Cheng Chang, Abstract Co-Author: Nothing to Disclose

Jyoji Nakagawara MD, Abstract Co-Author: Nothing to Disclose

Teiji Nakayama MD, Abstract Co-Author: Nothing to Disclose

Satoru Mori MD, Abstract Co-Author: Nothing to Disclose

Ryuji Sakakibara MD, Abstract Co-Author: Nothing to Disclose

00030490-DMT et al, Abstract Co-Author: Nothing to Disclose

Idiopathic normal pressure hydrocephalus (iNPH) has been shown to be treatable by shunt operation; however, since preoperative prediction of the effectiveness of shunt operation remains difficult, and postoperative complications occur in some patients, preoperative evaluation of the effectiveness of shunt operation and strict selection of patients with iNPH are still required. Using structural equation modeling (SEM) based on brain perfusion SPECT study, we tried to predict the degree of cognitive improvement following shunt operation in iNPH patients.

This study was performed in 79 patients with iNPH who were enrolled in a multicenter prospective study: study of idiopathic normal-pressure hydrocephalus on neurological improvement (SINPHONI). The correlations between pre-shunt and post-shunt Mini-Mental State Examination (MMSE) scores and the degrees of pre-shunt regional cerebral blood flow (CBF) reduction measured by N-isopropyl-p- [123I] iodoamphetamine SPECT were examined. Correlations with the regional CBF reduction degree (P<0.05) were regarded as parameters that affected MMSE scores, and SEM was performed to predict post-shunt cognitive improvement.

A significant negative correlation was observed between post-shunt MMSE scores and the degree of regional CBF reduction in the left prefrontal lobe, right prefrontal lobe, right parietal lobe, left posterior temporal lobe, and left medial frontal lobe. SEM for predicting post-shunt MMSE scores was developed using data on these regional CBF reduction degrees as parameters. The values of fitness and various statistical parameters of some structural equation models were calculated, and the model with four parameters: the left prefrontal, left posterior temporal, right parietal, and left medial frontal CBF reduction degrees, was selected. This model showed the best statistical fitness among the models. This model was used as a predictive equation and showed over 80% accuracy for predicting post-shunt MMSE scores.

To predict the effectiveness of the shunt operation for iNPH, post-shunt MMSE score changes were simulated by SEM using data on the degrees of pre-shunt regional CBF reduction, and satisfactory results were obtained.

Structural equation modeling will be a useful method to predict the effectiveness of shunt operation and to facilitate the selection of surgical candidates among patients with iNPH.

Goto, H, Ishii, K, Kazui, H, Hashimoto, M, Hayashida, K, Hashikawa, K, Chang, C, Nakagawara, J, Nakayama, T, Mori, S, Sakakibara, R, et al, 0, Predicative Value of Brain Perfusion SPECT for Cognitive Improvement with Shunt Operation in Idiopathic Normal Pressure Hydrocephalus (iNPH).  Radiological Society of North America 2009 Scientific Assembly and Annual Meeting, November 29 - December 4, 2009 ,Chicago IL. http://archive.rsna.org/2009/8003896.html