Goodarzi Zahra, Mrklas Kelly J, Roberts Derek J, Jette Nathalie, Pringsheim Tamara, Holroyd-Leduc Jayna
From the Departments of Community Health Sciences (Z.G., K.J.M., D.J.R., N.J., T.P., J.H.-L.) and Clinical Neurosciences (N.J., T.P.), University of Calgary; Research Priorities and Implementation (K.J.M.), Alberta Health Services, Edmonton; Departments of Critical Care Medicine (D.J.R.) and Surgery (D.J.R.), Hotchkiss Brain Institute and O'Brien Institute for Public Health (N.J., T.P., J.H.-L.), and Departments of Psychiatry and Pediatrics (T.P.) and Medicine (Z.G., J.H.-L.), University of Calgary and Alberta Health Services, Canada.
Neurology. 2016 Jul 26;87(4):426-37. doi: 10.1212/WNL.0000000000002898. Epub 2016 Jun 29.
Failure to detect depression in patients with Parkinson disease (PD) can lead to worsened outcomes for patients and caregivers. Accurate identification of depression would enable practitioners to provide comprehensive care for their patients with PD.
Our objective was to examine the diagnostic accuracy of tools for detecting depression in adult outpatients with PD. We searched MEDLINE, PsycINFO, and EMBASE (inception to December 1, 2015), gray literature, and bibliographies of included studies. The pooled prevalence of depression across studies and diagnostic accuracy estimates were calculated using random-effects models. Diagnostic accuracy estimates were calculated across the best-reported cutoffs from each study and across specific cutoffs, when feasible.
Out of 8,184 citations, 21 studies were included, evaluating 24 tools, with 4 amenable to meta-analysis. The pooled prevalence of major depression was 22.9% (95% confidence interval [CI] 18.1-27.7). The 15-item Geriatric Depression Scale (GDS-15) had a pooled sensitivity of 0.81 (95% CI 0.64-0.91) and specificity of 0.91 (95% CI 0.87-0.94). The most sensitive cutoff for the GDS-15 was 5 at 0.91 (95% CI 0.83-1.00). The Beck Depression Inventory I/Ia had a pooled sensitivity of 0.79 (95% CI 0.61-0.90) and specificity of 0.85 (95% CI 0.79-0.90). The Montgomery-Åsberg Depression Rating Scale yielded a pooled sensitivity of 0.77 (95% CI 0.69-0.83) and specificity of 0.92 (95% CI 0.79-0.97). The Unified Parkinson's Disease Rating Scale had a pooled sensitivity of 0.72 (95% CI 0.64-0.79) and specificity of 0.80 (95% CI 0.70-0.87). All estimates had heterogeneity.
There are several valid tools for detecting depression in patients with PD. Practitioners should choose one that fits their clinical practice.
未能在帕金森病（PD）患者中检测出抑郁症会导致患者及其照料者的预后恶化。准确识别抑郁症将使医生能够为PD患者提供全面的护理。
我们的目的是检验用于检测成年PD门诊患者抑郁症的工具的诊断准确性。我们检索了MEDLINE、PsycINFO和EMBASE（从创刊至2015年12月1日）、灰色文献以及纳入研究的参考文献。使用随机效应模型计算各研究中抑郁症的合并患病率和诊断准确性估计值。在可行的情况下，根据每项研究报告的最佳临界值以及特定临界值计算诊断准确性估计值。
在8184条文献中，纳入了21项研究，评估了24种工具，其中4种适合进行荟萃分析。重度抑郁症的合并患病率为22.9%（95%置信区间[CI]18.1 - 27.7）。15项老年抑郁量表（GDS - 15）的合并敏感性为0.81（95%CI 0.64 - 0.91），特异性为0.91（95%CI 0.87 - 0.94）。GDS - 15最敏感的临界值为5，敏感性为0.91（95%CI 0.83 - 1.00）。贝克抑郁量表I/Ia的合并敏感性为0.79（95%CI 0.61 - 0.90），特异性为0.85（95%CI 0.79 - 0.90）。蒙哥马利 - 阿斯伯格抑郁评定量表的合并敏感性为0.77（95%CI 0.69 - 0.83），特异性为0.92（95%CI 0.79 - 0.97）。统一帕金森病评定量表的合并敏感性为0.72（95%CI 0.64 - 0.79），特异性为0.80（95%CI 0.70 - 0.87）。所有估计值均存在异质性。
有几种有效的工具可用于检测PD患者的抑郁症。医生应选择适合其临床实践的工具。
