Gait disorder as a predictor of spatial learning and memory impairment in aged mice

OBJECTIVE To investigate whether gait dysfunction is a predictor of severe spatial learning and memory impairment in aged mice. METHODS A total of 100 12-month-old male mice that had no obvious abnormal motor ability and whose Morris water maze performances were not significantly different from those of two-month-old male mice were selected for the study. The selected aged mice were then divided into abnormal or normal gait groups according to the results from the quantitative gait assessment. Gaits of aged mice were defined as abnormal when the values of quantitative gait parameters were two standard deviations (SD) lower or higher than those of 2-month-old male mice. Gait parameters included stride length, variability of stride length, base of support, cadence, and average speed. After nine months, mice exhibiting severe spatial learning and memory impairment were separated from mice with mild or no cognitive dysfunction. The rate of severe spatial learning and memory impairment in the abnormal and normal gait groups was tested by a chi-square test and the correlation between gait dysfunction and decline in cognitive function was tested using a diagnostic test. RESULTS The 12-month-old aged mice were divided into a normal gait group (n = 75) and an abnormal gait group (n = 25). Nine months later, three mice in the normal gait group and two mice in the abnormal gait group had died. The remaining mice were subjected to the Morris water maze again, and 17 out of 23 mice in the abnormal gait group had developed severe spatial learning and memory impairment, including six with stride length deficits, 15 with coefficient of variation (CV) in stride length, two with base of support (BOS) deficits, five with cadence dysfunction, and six with average speed deficits. In contrast, only 15 out of 72 mice in the normal gait group developed severe spatial learning and memory impairment. The rate of severe spatial learning and memory impairment was significantly higher in the abnormal gait group as compared to that in the normal gait group (x = 21.986, P < 0.001). All five parameters used to assess gait predicted severe spatial learning and memory impairment in aged mice (P < 0.01). However, the difference of the area under the ROC (receiver operating characteristic) curve for each quantitative gait parameter was not statistically significant. CONCLUSION Gait disorders are a predictor of severe spatial learning and memory impairment in aged mice, and stride length, variability of stride length, base of support, cadence, and average speed are all sensitive parameters for assessing gait.