背景與目的：以頭部旋轉位置下之平滑追視（smooth pursuit）的速度來診斷頸因性暈眩，其敏感度和特異性分別可達90％和91 %，一般以之為頸因性暈眩的客觀評估工具。但此項檢查之信度尚未有文獻報告。本研究測量在頸部不同旋轉位置下的眼球追視速度之測量信度。方法：10位20－30歲的健康女性受試者，在相隔一週內進行兩次測試。以眼動分析系統，測試分為四種水平方向及三種垂直方向；包括頸部正中位置下右半視野內水平追視、頸部正中位置下左半視野內水平追視、頸部於左轉45度位置下左半視野內水平追視、及頸部於右轉45度位置下右半視野內水平追視、頸部正中位置下垂直追視、頸部於右轉45度下垂直追視、頸部於左轉45度下垂直追視七種。統計方法以無母數分析（Wilcoxon test）比較兩次測量之差異是否具統計意義。結果：水平方向眼球追視時，眼球運動速度略小於光點目標物運動速度，比值範圍為0.84～0.99；垂直方向眼球追視時，眼球運動速度均大於光點目標物運動速度，比值範圍為1.10～1.47。以無母數統計分析方法（Wilcoxon test）比較七種情況下之眼球運動速度比值，第一次與第二次測量之值間，除頸部於正中位置下之垂直向下追視有統計差異（p=0.028 ) ，其餘項目無統計差異（p>0.05)。結論：水平方向之眼球追視運動之測量信度良好。垂直眼球追視運動之影響需進一步控制眼瞼動作之干擾。（物理治療 2004 ; 29 ( 4 ) : 248－253)

Back ground and purposes: The velocity of smooth pursuit eye movement during neck pre-rotated position has been used to differential diagnosis the patients with cervicogenic vertigo due to traumatic neck injury. The aim of this investigation is to examine the test- retest reliability of the smooth pursuit eye movements during different head position. Methods: Ten asymptomatic young college students participated in this study (aged 20-30, female, right handed) to receive tests and were retested in one week. The horizontal smooth pursuit tests included eye movements in four conditions: in right (HRN) or left (HLN) visual field at neutral head position or at right (HRR) or left (HLL) pre-rotated head position. The vertical smooth pursuit tests included vertical smooth pursuit in three conditions: at neutral head position (VN), at right (VR) or left (VL) pre-rotated head position. The ratio of the velocity of eye movement and the reliability in these seven conditions were examined to test their reliabilities by using non-parametric Wilcoxon test. Results: The ratios of the eye velocity and target velocity during horizontal smooth pursuit were in the range of 0.84-0.99. During vertical smooth pursuit, the ratios of eye movement and target movement were in the range of 1.10-1.47. The test-retest velocity ratio of each smooth pursuit test was not significant different (p>0.05), except during downward eye movement in VN condition (p=0.028). Conclusions: The majority of the smooth pursuit test in the asymptomatic young subjects were reliable. However, further control of the motion of eye lid during vertical movement is needed to eliminate the measurement error. (FJPT 2004;29(4):248-253)