Reduced Pineal Volume in Alzheimer Disease: A Retrospective Cross-sectional MR Imaging Study

Purpose: To evaluate pineal volume in patients with Alzheimer disease (AD), patients with mild cognitive impairment (MCI), and healthy control subjects and to correlate the findings with results of cognitive testing and brain parenchymal volumes. Materials and Methods: The ethics committee approved this retrospective study. The participants included 63 patients with AD, 33 patients with MCI, and 24 healthy control subjects. There were 36 men and 84 women, with a mean age (+/- standard deviation) of 76.7 years +/- 7.6. The pineal gland volume and pineal parenchymal volume were measured by using three-dimensional volumetric magnetic resonance imaging (T1-weighted magnetization-prepared rapid gradient-echo sequence; spatial resolution, 0.9 X 0.98 X 0.98 mm). With age and total intracranial volume as covariates, analysis of covariance with the Bonferroni post hoc test was performed to compare the pineal volume among the AD, MCI, and control groups. Multiple regression analyses were used to identify predictor variables associated with pineal volume. Results: The mean pineal gland volume in patients with AD (72.3 mm(3) +/- 5.4; 95% confidence interval [CI]: 61.5 mm(3), 83.1 mm(3)) was significantly smaller than that in control subjects (102.1 mm(3) +/- 9.0; 95% CI: 84.4 mm(3), 119.9 mm(3)) (P = .019). The mean pineal parenchymal volume in patients with AD (63.8 mm(3) +/- 4.2; 95% CI: 55.4 mm(3), 72.1 mm(3)) was significantly smaller than that in patients with MCI (81.7 mm(3) +/- 5.8; 95% CI: 70.3 mm(3), 93.1 mm(3); P = .044) and control subjects (89.1 mm(3) +/- 6.9; 95% CI: 75.4 mm(3), 102.9 mm(3); P = .009). Multiple regression analyses demonstrated that the Mini-Mental State Examination score and total intracranial volume were significant independent predictors of both pineal gland volume and pineal parenchymal volume (P < .001). Conclusion: Pineal volume reduction showed correlation with cognitive decline and thus might be useful to predict cognitive decline in patients with AD. (C) RSNA, 2017