静息态功能磁共振成像联合神经黑色素成像对帕金森病的诊断价值

Resting-state rs-fMRI combined with NM-MRI in diagnosing Parkinson's disease

摘要:

目的 评估静息态功能磁共振磁共振成像（rs-fMRI）联合神经黑色素成像（NM-MRI）对帕金森病（PD）患者的诊断价值。方法 回顾性选取2022年1月—2023年12月淮安市第一人民医院收治的85例疑似PD患者，依据临床综合诊断结果分为PD组（45例）与非PD组（40例）。比较两组的临床特征（年龄、性别、神经系统症状、基础疾病史等）、静息态rs-fMRI指标［双侧小脑、左侧中央前回、丘脑、桥脑、双侧额叶、左前扣带回、颞叶、左岛叶的低频振幅（ALFF）值］及NM-MRI指标［黑质（SN）相对对比度、SN体积、SN校正后体积、蓝斑（LC）相对对比度］；采用多因素逐步Logistic回归模型分析PD的独立危险因素，绘制受试者工作特征（ROC）曲线评估各指标诊断PD的效能。结果 两组性别构成、年龄、神经系统症状、高血压史、糖尿病史、高脂血症、颞叶ALFF值和左岛叶ALFF值比较，差异均无统计学意义（P >0.05）。PD组双侧小脑ALFF值高于非PD组（P <0.05），左侧中央前回、丘脑、桥脑、双侧额叶、左前扣带回ALFF值均低于非PD组（P <0.05），PD组SN相对对比度、SN体积、SN校正后体积和LC相对对比度均低于非PD组（P <0.05）。多因素逐步Logistic回归分析结果显示：桥脑ALFF值高［O^R=0.049（95% CI：0.003，0.911）］、双侧额叶ALFF值高［O^R=0.071（95% CI：0.007，0.700）］、SN相对对比度高［O^R=0.767（95% CI：0.638，0.923）］、SN体积大［O^R=0.921（95% CI：0.863，0.983）］均为PD发生的保护因素（P <0.05）。桥脑ALFF值、双侧额叶ALFF值、SN相对对比度、SN体积联合诊断PD时，曲线下面积最大，为0.981，敏感性最高，为93.3%（95% CI：0.817，0.986），特异性最高，为95.0%（95% CI：0.831，0.994）。结论 静息态rs-fMRI的脑区ALFF值与NM-MRI的SN相关指标可作为诊断PD的有效影像学标志物，两者联合应用能显著提升诊断PD准确性，具有重要临床推广价值。

Abstract:

Objective To evaluate the value of resting-state functional magnetic resonance imaging (rs-fMRI) combined with neuromelanin MRI (NM-MRI) in diagnosing Parkinson's disease (PD).Methods Eighty-five patients with suspected PD admitted to Huai'an First People's Hospital from January 2022 to December 2023 were retrospectively enrolled and divided into a PD group (45 cases) and a non-PD group (40 cases) by comprehensive clinical diagnosis. We compared clinical characteristics, resting-state rs-fMRI indicators [amplitude of low-frequency fluctuation (ALFF) of bilateral cerebellum, left precentral gyrus, thalamus, pons, bilateral frontal lobes, left anterior cingulate gyrus, temporal lobe and left insula] and NM-MRI indicators [substantia nigra (SN) relative contrast, SN volume, corrected SN volume, locus coeruleus (LC) relative contrast] between the two groups. Multivariable logistic regression analysis was used to identify independent factors associated with PD, and receiver operating characteristic (ROC) curve analysis was used to assess diagnostic performance.Results Univariate analysis showed no significant differences between the PD group and the non-PD group in age, sex, neurological symptoms, histories of hypertension, diabetes mellitus, or dyslipidemia, or ALFF values in the temporal lobe and left insula (P > 0.05). The PD group exhibited significantly higher ALFF values in the bilateral cerebellum, but significantly lower ALFF values in the left precentral gyrus, thalamus, pons, bilateral frontal lobes, and left anterior cingulate gyrus, as well as significantly lower SN relative contrast, SN volume, corrected SN volume, and LC relative contrast compared with the non-PD group (P < 0.05). Multivariable logistic regression analysis revealed that pontine ALFF [O^R = 0.049 (95% CI: 0.003, 0.911) ], bilateral frontal ALFF [O^R = 0.071 (95% CI: 0.007, 0.700) ], SN relative contrast [O^R = 0.767 (95% CI: 0.638, 0.923) ], and SN volume [O^R = 0.921 (95% CI: 0.863, 0.983) ] were independent protective factors for PD (all P < 0.05). ROC curve analysis demonstrated that the combined use of the above four indicators yielded an area under the curve of 0.981 for diagnosing PD, with a sensitivity of 93.3% (95% CI: 0.817, 0.986) and a specificity of 95.0% (95% CI: 0.831, 0.994).Conclusion Regional ALFF values on resting-state rs-fMRI and SN-related NM-MRI indicators are effective imaging markers for PD. Their combination significantly improves diagnostic accuracy, with important clinical value.