Parkinsonism is a frequently encountered clinical feature in patients with bipolar disorder (BD). It is usually attributed to side effects of medication, but can also be a result of concomitant cerebrovascular disease and even an emerging idiopathic Parkinson's disease (PD). In a recent meta‐analysis by Faustino et al published in JAMA Neurology, the association of BD with a later diagnosis of PD was assessed.1 Four cohort studies and three cross‐sectional studies reporting data on the likelihood of developing PD in BD vs non‐BD populations were included, with a total of 4.374.211 participants overall. Two of the seven studies were considered to have an elevated risk of bias: one due to inclusion of a specific subgroup of veterans limiting comparability and one due to unreliable data collection (self‐reported diagnoses of PD) and a different source of the control group resulting in limited selection and comparability. In the meta‐analysis excluding these two, the likelihood of a subsequent diagnosis of idiopathic PD in patients previously diagnosed with BD was increased (odds ratio, 3.21; 95% CI, 1.89‐5.45; I² = 94%). Previous studies have suggested that PD is probably more common in BD than in the general population, but this review provides for the first time evidence supporting this notion based on a systematically analyzed, large pooled data set. The strength of this study is that a large number of subjects were included. However, several limitations are to be noted. It is not clear how PD‐diagnoses were established, most data were derived from medical records, both cross‐sectional and longitudinal (retrospective and prospective) data were included in the meta‐analysis leading to variation in data quality. Overestimation of the presence of PD is possible, as the diagnostic coding may not have differentiated PD from other causes of parkinsonism, and cases of atypical parkinsonism may have been included. Furthermore, the age of onset of BD in relation to the development of PD over time is not clear, therefore PD cases debuting with mood symptoms may have been included as patients with BD that developed PD subsequently. In a subgroup analysis longer follow‐up was associated with smaller increase in the risk of PD diagnosis. Bearing in mind the difference in the expected age at onset of BD (around 25 years) and PD (around 55 years) this is remarkable and requires further exploration to see if mostly young subjects have been followed longer or that specifically older late‐onset BD cases have an increased risk for PD.

All together the review and meta‐analysis suggest that patients with BD have a significantly increased risk of developing PD compared with the general population.1

This study has both clinical and research implications for BD and PD.

The identification of the underlying cause of parkinsonism in patients with BD is of significant clinical relevance. It is a challenge to determine the underlying cause of parkinsonism (defined by resting tremor, bradykinesia, rigidity and/or postural instability) in BD clinically. Clinical signs such as asymmetric motor features, the presence of resting tremor and hyposmia are less suggestive for drug‐induced parkinsonism, but lack specificity.2 Additional imaging of the brain with Computer Tomography or Magnestic Resonance Imaging could rule out cerebrovascular pathology. The most reliable way to differentiate PD from other causes of parkinsonism is dopamine transporter (DAT) imaging by ¹²³I‐N‐ω‐fluoropropyl‐2β‐carbomethoxy‐3β‐(4‐iodophenyl) nortropane (¹²³I‐FP‐CIT) single photon emission computed tomography (SPECT) as a proxy of nigrostriatal degeneration. An ¹²³I‐FP‐CIT scan is generally normal in drug‐induced or vascular parkinsonism.

In patients with BD, longitudinal motor assessments could be recommended to screen for prodromal motor and nonmotor signs of PD. Once patients with BD present with tremor or other symptoms of parkinsonism this cannot simply be attributed as drug‐induced and investigation of PD is recommended. Preferably this should be done by a neurologist and include dopaminergic imaging when in doubt. If these symptoms are a result of side effects, these are often dose‐dependent and lowering lithium levels may reduce symptoms of parkinsonism without losing efficacy, especially in older adults.2, 3

Probably not all BD patients in which parkinsonism becomes manifest should be assessed for PD promptly. The meta‐analysis suggests that only a small proportion of the BD patients are at risk for developing subsequent PD. Next step would be to investigate which patients are at risk and how they can be identified. It seems logical that age is a factor that should be taken into account since PD usually manifests at older age. Also risk estimates per gender should be investigated since PD is more predominant in males.

Of course, this raises the question of the mechanism behind the link between BD and PD. BD can be seen as a brain disorder that debuts earlier in life and converts to neurodegenerative disorder with ageing. Risk factors for this conversion could be genetic or environmental (head trauma, toxic exposure or unfavorable course of BD with allostatic loading by multiple episodes, substance abuse and psychiatric comorbidities). The authors quote the dopamine dysregulation hypothesis that links BD to PD: manic states may lead to a downregulation of dopamine receptor sensitivity resulting in a depressive episode that is in turn compensated by upregulation resulting in a (hypo)manic episode. Numerous cycles may lead to an overall reduction of dopaminergic activity, the prototypical PD state. Further clinical evidence of dopamine dysregulation is found in the fact that patients with BD in a depressed mood experience more parkinsonism, and that patients with PD experience mood changes related to the on‐time/off‐time phenomena, with mania‐like symptoms in the on‐time phase and vice versa. Moreover, levodopa (dopamine agonist) can be used to increase dopamine levels as a therapeutic agent in PD but is also known to induce (hypo) mania, antipsychotic medication that block dopamine receptors can in turn improve manic symptoms but are also known to increase rigidity and hypokinesia in patients with PD.

A shared pathophysiological mechanism (causing dopamine dysregulation) for BD and PD could explain the overlap in a specific subset of patients. Whether lithium can prime for PD is unknown, but parkinsonism is a known side‐effect. However, dementia rates are lower in individuals exposed to standard or trace‐dose lithium, suggesting that neurodegeneration is not induced by lithium itself. Protective factors could also be identified, such as the use of lithium, known to inhibit excitatory neurotransmitters such as dopamine and to have neuroprotective effects.

Another hypothesis could be that BD is a not only a possible risk factor for developing PD later in life but could also be a prodromal feature of PD. In that light it is worth to perform risk‐analyses of BD in the Lewy body disease spectrum, including also Dementia with Lewy Bodies (DLB). Cognitive impairment is considered a common feature in BD, with a progressive nature in a subgroup leading to increased rates of dementia.4 The neurobiological underpinnings of this cognitive impairment remain unknown, but risk factors such as numerous mood episodes, psychiatric admissions, psychotic features, earlier age at onset of BD and cardiovascular burden have been suggested.5 DLB is defined by progressive cognitive decline accompanied by parkinsonism, visual hallucinations, cognitive fluctuations, and RBD. There is a large overlap with PD both clinically and pathologically. In contrast to Alzheimer's dementia (AD), memory dysfunction is not prominent and brain‐atrophy on structural imaging is relatively absent in DLB, often resulting is a missed diagnosis of DLB in patients with BD presenting increased mood instability and cognitive decline later in life. It is highly likely that similar risk scores as found in relation to PD could be found in DLB. It is known that pathological processes of neurodegenerative diseases like PD, DLB and AD start decades before clinical symptoms manifest. However, it is increasingly recognized that nonmotor, noncognitive phenomena already present in early stages, but usually at that point are not attributed to neurodegenerative disease. A well‐studied syndrome is REM‐Sleep behavior disorder (RBD). This syndrome usually manifests before midlife. Over 80% of the patients with RBD develop Lewy body disease (PD or DLB) and the syndrome is considered a prodromal sign of these diseases. Further research should be undertaken to investigate the possibility if a subgroup of BD could be at a prodromal stage of Lewy Body Diseases. This would have implications for future treatment strategies when potential disease modifying drugs for Lewy body diseases become available. Further pathophysiological and genetic studies should shed light on biological mechanisms that link BD and Lewy body diseases. Both disease entities are known to have genetic susceptibility and possibly genetic variants could explain the co‐existence of both diseases.

This review on the increased risk for PD in patients with BD urges for a new clinical approach. Tremor or other symptoms of parkinsonism in patients with BD cannot simply be attributed as drug‐induced and investigation of PD is recommended including dopaminergic imaging, especially in older and late‐onset cases.

As for research, BD can be seen as a neuropsychiatric disorder that debuts earlier in life with mood symptoms and associated motor and cognitive symptoms later in life may be the first symptoms of a neurodegenerative course. Risk factors for this conversion could be genetic or environmental.

A close collaboration between neurologists and psychiatrists may open new avenues for management of both BD and PD and research in the neurobiological underpinnings of BD.

帕金森病是双相情感障碍（BD）患者的常见临床特征。它通常归因于药物的副作用，但也可能是伴随性脑血管疾病甚至新兴的特发性帕金森氏病（PD）的结果。在Faustino等人最近发表在《 JAMA Neurology》上的荟萃分析中，评估了BD与后来的PD诊断的关联。1个包括四项队列研究和三项横断面研究，这些研究报告了BD和非BD人群中PD发生可能性的数据，总共有4.3742.211名参与者。七项研究中的两项被认为具有较高的偏见风险：一项是由于纳入了特定的退伍军人亚组而限制了可比性，另一项是由于数据收集不可靠（PD的自我报告诊断）以及对照组的来源不同导致选择和可比性受到限制。在排除这两个因素的荟萃分析中，先前被诊断为BD的患者随后诊断为特发性PD的可能性增加了（优势比为3.21； 95％CI为1.89-5.45； I ² = 94％）。先前的研究表明，PD可能在BD中比在一般人群中更为常见，但是该综述首次基于系统分析的大型汇总数据集首次提供了支持该观点的证据。这项研究的优势在于包括了大量的主题。但是，要注意一些限制。目前尚不清楚如何进行PD诊断，大多数数据来自病历，荟萃分析中同时包括横断面和纵向（回顾性和前瞻性）数据，从而导致数据质量的差异。PD的存在可能被高估，因为诊断编码可能无法将PD与其他帕金森病原因区分开，并且可能包括非典型帕金森病病例。此外，BD的发病年龄与PD随时间的发展有关尚不清楚，因此，初次出现情绪症状的PD病例可能被包括为随后发展PD的BD患者。在亚组分析中，较长的随访与PD诊断风险增加较小相关。考虑到BD发病的预期年龄（约25岁）和PD发病的预期年龄（约55岁）之间的差异是显着的，需要进一步研究以查看大多数年轻受试者的随访时间是否更长，或者具体而言是较晚发病的BD病例患PD的风险增加。在亚组分析中，较长的随访与PD诊断风险增加较小相关。考虑到BD发病的预期年龄（约25岁）和PD发病的预期年龄（约55岁）之间的差异是显着的，需要进一步研究以查看大多数年轻受试者的随访时间是否更长，或者具体而言是较晚发病的BD病例患PD的风险增加。在亚组分析中，较长的随访与PD诊断风险增加较小相关。考虑到BD发病的预期年龄（约25岁）和PD发病的预期年龄（约55岁）之间的差异是显着的，需要进一步研究以查看大多数年轻受试者的随访时间是否更长，或者具体而言是较晚发病的BD病例患PD的风险增加。

综上所述，回顾和荟萃分析表明，与普通人群相比，BD患者患PD的风险显着增加。1个

这项研究对BD和PD具有临床和研究意义。

BD患者帕金森病的根本原因的识别具有重要的临床意义。临床上确定BD帕金森病（由静息性震颤，运动迟缓，僵硬和/或姿势不稳定义）的根本原因是一项挑战。运动性不对称，静息性震颤和低渗等临床症状对药物诱发的帕金森综合症提示较少，但缺乏特异性。2使用计算机断层扫描或磁共振成像对大脑进行额外的成像可排除脑血管病变。通过从帕金森其他原因引起的分化PD的最可靠的方法是多巴胺转运蛋白（DAT）成像¹²³ I-N-ω氟丙基-2β甲酯基3β-（4-碘苯基）莨菪烷（¹²³I-FP-CIT）单光子发射计算机断层扫描（SPECT）作为黑纹状体变性的代表。的¹²³ I-FP-CIT扫描通常垂直于药物诱导或血管性帕金森综合征。

对于患有BD的患者，建议进行纵向运动评估以筛查PD的前驱运动和非运动征象。BD患者一旦出现震颤或其他帕金森氏症状，就不能简单地归因于药物诱导，建议进行PD研究。优选地，这应该由神经科医生完成，并且当有疑问时包括多巴胺能成像。如果这些症状是副作用的结果，则通常是剂量依赖性的，降低锂含量可以减轻帕金森综合征的症状而又不降低疗效，尤其是在老年人中。2 3

可能并非所有表现为帕金森氏症的BD患者都应立即进行PD评估。荟萃分析表明，只有一小部分BD患者有发展为继发PD的风险。下一步将是调查哪些患者处于危险之中，以及如何识别他们。似乎逻辑上应该考虑年龄因素，因为PD通常表现为年龄较大。另外，由于PD在男性中更为普遍，因此应调查每性别的风险估计。

当然，这提出了BD和PD之间的链接背后的机制问题。BD可以看作是一种脑部疾病，在一生中就较早出现，并随着年龄的增长而转变为神经退行性疾病。发生这种转变的危险因素可能是遗传或环境因素（头部外伤，中毒暴露或BD的病程不利，并伴有多发性变态反应，药物滥用和精神病合并症）。作者引用了将BD与PD关联的多巴胺失调假说：躁狂状态可能导致多巴胺受体敏感性下调，从而导致抑郁发作，而上调又弥补了上调导致的（低）躁狂发作。许多周期可能会导致多巴胺能活性（典型的PD状态）全面降低。多巴胺失调的进一步临床证据是，患有抑郁症的BD患者经历了更多的帕金森病，而患有PD的患者经历了与通/断时间现象相关的情绪变化，并在其中出现了躁狂症症状。准时阶段，反之亦然。此外，左旋多巴（多巴胺激动剂）可用于提高PD中的多巴胺水平，但也可引起（低）躁狂症；阻断多巴胺受体的抗精神病药可改善躁狂症状，但也已知会增加僵硬性和PD患者运动功能减退。

BD和PD的共同病理生理机制（引起多巴胺失调）可以解释特定患者亚组中的重叠。锂是否可以引发PD尚不清楚，但帕金森氏病是已知的副作用。但是，接触标准或痕量锂的人的痴呆症发生率较低，这表明锂本身并不会诱发神经变性。还可以确定保护因子，例如使用锂，已知该锂可抑制兴奋性神经递质（如多巴胺）并具有神经保护作用。

另一个假设可能是BD不仅是生命后期发展PD的可能危险因素，而且还可能是PD的前驱特征。因此，值得在路易体病谱中进行BD的风险分析，包括痴呆伴路易体（DLB）。认知障碍被认为是BD的常见特征，在亚组中具有渐进性，导致痴呆率增加。4这种认知障碍的神经生物学基础仍然是未知的，但是已经提出了危险因素，例如许多情绪发作，精神病入院，精神病特征，BD发病年龄较早和心血管负担。5DLB定义为进行性认知功能减退，伴有帕金森病，视觉幻觉，认知波动和RBD。在临床和病理上与PD有很大的重叠。与阿尔茨海默氏痴呆症（AD）相比，DLB的记忆障碍并不明显，并且在结构成像方面脑萎缩相对较少，通常导致BD患者缺乏DLB诊断，表现为情绪不稳定和生命后期认知下降。在DLB中很可能会发现与PD相关的相似风险评分。众所周知，PD，DLB和AD等神经退行性疾病的病理过程在临床症状显现之前就已经开始了数十年。但是，人们越来越认识到，在早期阶段就已经存在非运动，非认知现象，但通常在这一点上不归因于神经退行性疾病。一个被充分研究的综合症是REM睡眠行为障碍（RBD）。该综合征通常在中年之前表现出来。超过80％的RBD患者发展了路易体病（PD或DLB），该综合征被认为是这些疾病的前兆体征。应该进行进一步的研究，以调查BD亚组是否可能处于路易体病的前驱阶段。当路易氏体病的潜在疾病改良药物可用时，这将对未来的治疗策略产生影响。进一步的病理生理学和遗传学研究应阐明与BD和路易体病相关的生物学机制。

这项有关BD患者PD风险增加的综述呼吁寻求一种新的临床方法。BD患者的震颤或其他帕金森氏症状不能简单地归因于药物诱导，建议对PD进行检查，包括多巴胺能显像，尤其是在老年和晚期发病的患者中。

至于研究，BD可以被视为一种神经精神疾病，它在生命早期出现并伴有情绪症状，并在生命后期出现相关的运动和认知症状，可能是神经退行性疾病的最初症状。此转换的风险因素可能是遗传或环境因素。

神经科医生和精神科医生之间的密切合作可能会为BD和PD的管理以及BD神经生物学基础的研究开辟新的途径。