Abstract

BACKGROUND
Recently, there are many reports about the association of Diffusion Weighted Imaging (DWI) and the prognosis of hypoglycemic encephalopathy (HE), but those relationships have not yet been completely determined. As such, we researched for prognosis, according to a variety of clinical data, and the lesion's distribution on DWI.
METHODS
We retrospectively reviewed 19 patients who were diagnosed as HE. In addition, those prognoses were analyzed by a variety of clinical data and characteristics of lesion's distribution, which were evaluated on DWI and Apparent Diffusion Coefficient (ADC) maps. Three months later, those prognoses were determined by each Modified Rankin Scale. Further, the time-dependent average Glasgow Coma Scale (GCS), among the groups according to the characteristics of lesion's distributions in the initial DWI, was estimated.
RESULTS
In this study, the difference of prognosis was not shown, according to all the clinical data, such as the severity or duration of the hypoglycemic state, but the group that did not have any pathologic lesion on the initial DWI demonstrated a better prognosis, in comparison with the groups-that exhibited any positive lesion on the initial DWI (p = 0.006). The group that had a focal pathologic lesion on the initial DWI showed a better prognosis than the diffuse lesion's group (p = 0.045). The groups with no lesion or focal lesion showed a faster recovery of GCS than the other groups with a positive lesion or diffuse lesion within the initial 1 week.
CONCLUSIONS
We can identify that the characteristics of the lesion's distribution of DWI can be more helpful to predict of prognosis in HE than a variety of clinical data, such as the severity or duration of the hypoglycemic state.

• Keywords: diffusion weighted imaging;hypoglycemia;prognosis;