Jepsen Jørgen R, Laursen Lise H, Hagert Carl-Göran, Kreiner Svend, Larsen Anders I
Department of Occupational Medicine, Sydvestjysk Sygehus, Østergade 81-83, DK-6700 Esbjerg, Denmark.
BMC Neurol. 2006 Feb 27;6:10. doi: 10.1186/1471-2377-6-10.
In a sample of patients in clinical occupational medicine we have demonstrated that an upper limb neurological examination can reliably identify patterns of findings suggesting upper limb focal neuropathies. This further study aimed at approaching the diagnostic accuracy of the examination.
82 limbs were semi-quantitatively assessed by two blinded examiners (strength in 14 individual muscles, sensibility in 7 homonymous territories, and mechanosensitivity at 10 locations along nerves). Based on the topography of nerves and their muscular and sensory innervation we defined 10 neurological patterns each suggesting a localized nerve affliction. Information on complaints (pain, weakness and/or numbness/tingling) collected by others served as a reference for comparison. The relation between the presence of pattern(s) and complaints was assessed by kappa-statistics. Sensitivity, specificity, and positive/negative predictive values were calculated, and pre-test odds were compared to post-test probability.
The two examiners identified pattern(s) suggesting focal neuropathy in 34/36 out of 38 symptomatic limbs, respectively (kappa = 0.70/0.75), with agreement in 28 limbs. Out of 44 non-symptomatic limbs the examiners agreed on absence of any pattern in 38 limbs. With concordance between the examiners with regard to the presence or absence of any pattern, the sensitivity, specificity, positive and negative predictive values were 0.73, 0.86, 0.93 and 0.90, respectively. While the pre-test odds for a limb to be symptomatic amounted to 0.46 the post-test probability was 0.81. For each examiner the post-test probability was 0.87 and 0.88, respectively.
The improved diagnostic confidence is an indication of one aspect of construct validity of the physical examination. For determination of clinical feasibility of the examination further studies are required, most importantly 1) studies of validity by means of comparison with additional references and 2) studies of the potential benefit that can be attained from its use.
在临床职业医学患者样本中,我们已证明上肢神经学检查能够可靠地识别提示上肢局灶性神经病变的检查结果模式。本进一步研究旨在探讨该检查的诊断准确性。
由两名盲法检查者对82条肢体进行半定量评估(14块单独肌肉的力量、7个同名区域的感觉以及沿神经10个位置的机械敏感性)。根据神经的解剖位置及其肌肉和感觉支配,我们定义了10种神经学模式,每种模式提示局部神经病变。由其他人收集的关于主诉(疼痛、无力和/或麻木/刺痛)的信息用作比较的参考。通过kappa统计评估模式的存在与主诉之间的关系。计算敏感性、特异性以及阳性/阴性预测值,并比较检验前概率与检验后概率。
两名检查者分别在38条有症状肢体中的34/36条中识别出提示局灶性神经病变的模式(kappa = 0.70/0.75),在28条肢体上达成一致。在44条无症状肢体中,检查者在38条肢体上一致认为没有任何模式。在检查者关于是否存在任何模式达成一致的情况下,敏感性、特异性、阳性和阴性预测值分别为0.73、0.86、0.93和0.90。虽然一条肢体有症状的检验前概率为0.46,但检验后概率为0.81。对于每位检查者,检验后概率分别为0.87和0.88。
诊断信心的提高表明体格检查在结构效度方面的一个方面。为了确定该检查的临床可行性,需要进一步研究,最重要的是1)通过与其他参考方法比较进行效度研究,以及2)研究使用该检查可能获得的潜在益处。