Kaiser Permanente Division of Research, Oakland, California, USA.
Department of Radiological Sciences, University of California Irvine School of Medicine, Irvine, California, USA.
J Womens Health (Larchmt). 2021 Jun;30(6):848-856. doi: 10.1089/jwh.2020.8372. Epub 2020 Dec 7.
Since vascular risk factors are implicated in cognitive decline, and breast arterial calcification (BAC) is related to vascular risk, we postulated that BAC may be associated with cognitive impairment and dementia. We used a multiethnic cohort of 3,913 asymptomatic women 60-79 years of age recruited after mammography screening at a large health plan in 2012-2015. A BAC mass score (mg) was derived from digital mammograms. Cognitive function was measured at baseline using the Montreal Cognitive Assessment (MoCA) and incident all-cause dementia ( = 49 events; median follow-up = 5.6 years) were ascertained with validated ICD-9 and ICD-10 codes. We used cross-sectional linear regression of MoCA scores on BAC, then multinomial logistic regression predicting mild cognitive impairment not progressing to dementia and incident all-cause dementia and, finally, Cox regression of incident all-cause dementia. No association by linear regression was found between MoCA scores and BAC presence in unadjusted or adjusted analysis. Women with severe (upper tertile) BAC had a MoCA score lower by 0.58 points (standard error [SE] = 0.18) relative to women with no BAC. However, this difference disappeared after multivariate adjustment. No significant associations were found in multinomial logistic regression for either BAC presence or gradation in unadjusted or adjusted analysis. No significant associations were found between BAC presence with incident all-cause dementia (fully adjusted hazard ratio = 0.74; 95% confidence interval: 0.39-1.39). Likewise, no significant association with incident all-cause dementia was noted for BAC gradation. Our results do not support the hypothesis that BAC presence or gradation may contribute to cognitive impairment or development of all-cause dementia.
由于血管危险因素与认知能力下降有关，而乳腺动脉钙化（BAC）与血管风险有关，我们假设 BAC 可能与认知障碍和痴呆有关。我们使用了一个多民族队列，其中包括 3913 名无症状的 60-79 岁女性，这些女性是在 2012-2015 年的一次大型健康计划中的乳房 X 光筛查后招募的。从数字乳房 X 光片中得出 BAC 质量评分（mg）。在基线时使用蒙特利尔认知评估（MoCA）测量认知功能，并使用经过验证的 ICD-9 和 ICD-10 代码确定所有原因痴呆的发病情况（ = 49 例事件；中位随访时间 = 5.6 年）。我们使用 MoCA 评分与 BAC 的横截面线性回归，然后使用多分类逻辑回归预测从轻度认知障碍到痴呆的进展以及所有原因痴呆的发病情况，最后使用 Cox 回归分析所有原因痴呆的发病情况。在未经调整或调整后的分析中，MoCA 评分与 BAC 的存在之间没有线性回归关系。与没有 BAC 的女性相比，BAC 严重（上三分位数）的女性 MoCA 评分低 0.58 分（标准误差 [SE] = 0.18）。然而，这种差异在多变量调整后消失了。在未经调整或调整后的分析中，多分类逻辑回归对于 BAC 的存在或分级均未发现显著相关性。在 BAC 存在与所有原因痴呆的发病风险之间未发现显著相关性（完全调整后的危害比 = 0.74；95%置信区间：0.39-1.39）。同样，BAC 分级与所有原因痴呆的发病风险也没有显著相关性。我们的研究结果不支持 BAC 的存在或分级可能导致认知障碍或全因痴呆的假设。
