Detsky Michael E, McDonald Devon R, Baerlocher Mark O, Tomlinson George A, McCrory Douglas C, Booth Christopher M
Faculty of Medicine, Medicine, University of Toronto, Toronto, Ontario.
JAMA. 2006 Sep 13;296(10):1274-83. doi: 10.1001/jama.296.10.1274.
In assessing the patient with headache, clinicians are often faced with 2 important questions: Is this headache a migraine? Does this patient require neuroimaging? The diagnosis of migraine can direct therapy, and information obtained from the history and physical examination is used by physicians to determine which patients require neuroimaging.
To determine the usefulness of the history and physical examination that distinguish patients with migraine from those with other headache types and that identify those patients who should undergo neuroimaging.
A systematic review was performed using articles from MEDLINE (1966-November 2005) that assessed the performance characteristics of screening questions in diagnosing migraine (with the International Headache Society diagnostic criteria as a gold standard) and addressed the accuracy of the clinical examination in predicting the presence of underlying intracranial pathology (with computed tomography/magnetic resonance imaging as the reference standard).
Two authors independently reviewed each study to determine eligibility, abstract data, and classify methodological quality using predetermined criteria. Disagreement was resolved by consensus with a third author.
Four studies of screening questions for migraine (n = 1745 patients) and 11 neuroimaging studies (n = 3725 patients) met inclusion criteria. All 4 of the migraine studies illustrated high sensitivity and specificity if 3 or 4 criteria were met. The best predictors can be summarized by the mnemonic POUNDing (Pulsating, duration of 4-72 hOurs, Unilateral, Nausea, Disabling). If 4 of the 5 criteria are met, the likelihood ratio (LR) for definite or possible migraine is 24 (95% confidence interval [CI], 1.5-388); if 3 are met, the LR is 3.5 (95% CI, 1.3-9.2), and if 2 or fewer are met, the LR is 0.41 (95% CI, 0.32-0.52). For the neuroimaging question, several clinical features were found on pooled analysis to predict the presence of a serious intracranial abnormality: cluster-type headache (LR, 10.7; 95% CI, 2.2-52); abnormal findings on neurologic examination (LR, 5.3; 95% CI, 2.4-12); undefined headache (ie, not cluster-, migraine-, or tension-type) (LR, 3.8; 95% CI, 2.0-7.1); headache with aura (LR, 3.2; 95% CI, 1.6-6.6); headache aggravated by exertion or a valsalva-like maneuver (LR, 2.3; 95% CI, 1.4-3.8); and headache with vomiting (LR, 1.8; 95% CI, 1.2-2.6). No clinical features were useful in ruling out significant pathologic conditions.
The presence of 4 simple historical features can accurately diagnose migraine. Several individual clinical features were found to be associated with a significant intracranial abnormality, and patients with these features should undergo neuroimaging.
在评估头痛患者时，临床医生常常面临两个重要问题：该头痛是否为偏头痛？该患者是否需要进行神经影像学检查？偏头痛的诊断可指导治疗，医生会根据病史和体格检查所获得的信息来确定哪些患者需要进行神经影像学检查。
确定病史和体格检查在区分偏头痛患者与其他头痛类型患者以及识别哪些患者应接受神经影像学检查方面的有用性。
采用来自MEDLINE（1966年 - 2005年11月）的文章进行系统评价，这些文章评估了筛查问题在诊断偏头痛（以国际头痛协会诊断标准为金标准）中的性能特征，并探讨了临床检查在预测潜在颅内病变存在方面的准确性（以计算机断层扫描/磁共振成像为参考标准）。
两位作者独立审查每项研究，以确定其是否符合纳入标准、提取数据，并使用预定标准对方法学质量进行分类。如有分歧，通过与第三位作者达成共识来解决。
四项关于偏头痛筛查问题的研究（n = 1745例患者）和十一项神经影像学研究（n = 3725例患者）符合纳入标准。如果满足3或4条标准，所有4项偏头痛研究均显示出高敏感性和特异性。最佳预测指标可用助记词POUNDing（搏动性、持续4 - 72小时、单侧性、恶心、致残性）概括。如果满足5条标准中的4条，确诊或可能为偏头痛的似然比（LR）为24（95%置信区间[CI]，1.5 - 388）；如果满足3条，LR为3.5（95% CI，1.3 - 9.2），如果满足2条或更少，LR为0.41（95% CI，0.32 - 0.52）。对于神经影像学问题，汇总分析发现几个临床特征可预测严重颅内异常的存在：丛集性头痛（LR，10.7；95% CI，2.2 - 52）；神经系统检查异常（LR，5.3；95% CI，2.4 - 12）；不明类型头痛（即非丛集性、偏头痛或紧张型）（LR，3.8；95% CI，2.0 - 7.1）；有先兆的头痛（LR，3.2；95% CI，1.6 - 6.6）；因用力或类似瓦尔萨尔瓦动作而加重的头痛（LR，2.3；95% CI，1.4 - 3.8）；伴有呕吐的头痛（LR，1.8；95% CI，1.2 - 2.6）。没有临床特征可用于排除重大病理状况。
存在4个简单的病史特征可准确诊断偏头痛。发现几个个体临床特征与重大颅内异常相关，具有这些特征的患者应接受神经影像学检查。
