比較催汗試驗及密西根神經病變篩檢量表對不同病程之第二型糖尿病病人周邊神經病變評估之一致性

背景及目的：糖尿病是全球普遍存在的慢性疾病，隨病程進展，會增加周邊神經病變進而導致潰瘍和截肢的發生。臨床上神經傳導檢查被視為診斷周邊神經病變的標準方法，但卻因費時且昂貴，無法普及於各級醫療機構。因此，能方便、無侵入和準確地早期檢測遠端小神經纖維神經病變的工具，極其重要。本研究以催汗試驗及密西根神經病變篩檢量表臨床檢查來比較不同病程第二型糖尿病周邊神經病變之一致性。研究方法：本研究採橫斷式研究設計（Cross-sectional study design），立意取樣263位台中某醫學中心糖尿病衛教中心病人為樣本。收集資料包括:糖尿病病程，人口學特性、生理量測、生活型態、臨床檢驗值、催汗試驗的電化學皮膚電導和密西根神經病變篩檢量表臨床檢查得分等。研究結果：結果有19%為密西根神經病變篩檢異常者；49%為催汗試驗檢查任一側腳掌異常者；64%為任一側手掌異常。而罹患糖尿病的病程長短與密西根神經病變篩檢量表異常、腳掌或手掌催汗試驗異常皆無顯著相關性。多變量羅吉斯迴歸分析結果發現：病程≦10年者，左手掌電化學皮膚電導每增加1 μS，發生密西根神經病變篩檢異常的機率是密西根神經病變篩檢量表正常的0.97倍(p=0.026)；病程＞10年者，左手掌電化學皮膚電導每增加1 μS，發生密西根神經病變篩檢異常的機率是密西根神經病變篩檢量表正常的0.89倍(p=0.043)。ROC曲線分析結果，病程＞10年者，右腳掌電化學皮膚電導為最佳預測密西根神經病變篩檢量表異常的相關因子。Cohen’s kappa分析結果發現：本研究樣本的雙腳掌催汗試驗和密西根神經病變篩檢量表的Kappa係數為0.184，雙手掌催汗試驗和密西根神經病變篩檢量表的Kappa係數為0.084；病程≦10年和病程＞10年的雙腳掌催汗試驗和密西根神經病變篩檢量表的Kappa係數分別為0.137和0.237(p＜0.05)。結論與建議：催汗試驗和密西根神經病變篩檢量表的一致性偏低，但以糖尿病的病程長短而言，雙腳掌催汗試驗仍可呼應密西根神經病變篩檢量表的結果。建議未來進行年度糖尿病神經病變篩檢時，可早期使用催汗試驗進行常規篩檢。

Background and Purpose: Diabetes is a chronic disease prevalent worldwide, and as the disease progresses, it increases peripheral neuropathy and leads to ulceration and amputation. Nerve Conduction Study is the standard clinical method for diagnosing peripheral neuropathy. Still, it is too time-consuming and expensive to be widely available at all levels of care. Therefore, tools that can be convenient, non-invasive, and accurate for the early detection of distal small nerve fiber neuropathy are essential. Therefore, this study aims to comparison of the consistency of the Sudomotor Testing and Michigan Neuropathy Screening Scale in the assessment of peripheral neuropathy in patients with Type 2 Diabetes Mellitus of different disease duration.Method：In this study, a cross-sectional study design was adopted, and 263 patients from the Diabetes Health Education Center of a medical center in Taichung were purpose sampled. The collected data included: diabetes mellitus history, demographic characteristics, physiological measurements, lifestyle patterns, clinical test data, Michigan Neuropathy Screening Scale and Sudomotor testing.Result：The results showed that 19% of the Michigan Neuropathy Screening Instrument was abnormal; 49% had abnormal foot ESC on either side by sudomotor test, and 64% had abnormal hand ESC on either side. Multivariable logistic regression analysis results found that: disease duration≦10 years: for every 1μS increase in left hands ESC, the probability of abnormal Michigan Neuropathy Screening Instrument examination is 0.97 times that of normal Michigan Neuropathy Screening Instrument examination (p=0.026); disease duration > 10 years: For every 1μS increase in left hands ESC, the probability of abnormal Michigan Neuropathy Screening Instrument examination is 0.89 times that of normal Michigan Neuropathy Screening Instrument examination (p=0.043). ROC curve analysis results show that in patients with disease duration >10 years, right foot ESC is the best predictor of abnormality in Michigan Neuropathy Screening Instrument examination. Cohen’s kappa analysis results found: The Kappa coefficient of foot ESC and Michigan Neuropathy Screening Instrument examination of the sample in this study is 0.184. The Kappa coefficient of hands ESC and Michigan Neuropathy Screening Instrument examination is 0.084; The Kappa coefficients of foot ESC and Michigan Neuropathy Screening Instrument examination for disease duration ≦10 years and disease duration >10 years were 0.137 and 0.237 respectively (p<0.05).conclusions and recommendations：The consistency between Sudomotor Testing and Michigan Neuropathy Screening Instrument examination is low. However, depending on the duration of the disease, foot ESC can still echo the results of the Michigan Neuropathy Screening Instrument examination. It is recommended that Sudomotor Testing be used early in routine screening for annual diabetic neuropathy screening in the future.