探討Sniffin’ sticks identification在第五期慢性腎臟病病人的短期再測信度

Abstract

研究背景與目的: 慢性腎臟病病人由於體內毒素累積，使嗅上皮及嗅球的內在抗氧化防禦系統遭破壞，以致其嗅覺功能劣於正常人，過去文獻指出經血液透析治療後可改善腎臟病病人的嗅覺辨識度，但由於過去在Sniffin’ sticks測量工具上未曾檢查其短期的再測信度，無法確認嗅覺辨識度改善是否來自工具的測量誤差，因此本研究目的除探討第五期慢性腎臟病病人嗅覺辨識度的現況外，欲測試Sniffin’ sticks 嗅覺辨識度短期的再測信度並計算最小可偵測變化值。 研究方法: 本研究為縱慣性觀察性研究，自2020年5月1日至 2020 年 10 月 31 日，以台灣大學附設醫院腎臟科門診，尚未接受透析治療的第五期慢性腎臟病病人且蒙特婁認知量表分數大於23.5分者為對象，所有受試者以Sniffin’ sticks在間隔1-2小時內重複兩次嗅覺辨識度測試，考量短時間內重複測量的學習性效應，故打亂嗅味棒測試順序並確定前後兩次測驗中的氣味棒順序不同。 研究結果: 本研究共納入60位尚未接受透析治療的第五期慢性腎臟病病人，平均年齡為63.6± 14.1歲，75%的受試者教育程度為高中以上，其中有嗅覺辨識度異常的盛行率為20%，受試者的年齡、性別、有無糖尿病、腎絲球過濾率、血中尿素氮及肌肝酸濃度與嗅覺辨識度異常的比率無關。Sniffin’ sticks嗅覺辨識度測試短期再測信度上，前後兩次嗅覺辨識度測試的組內相關係數ICC為0.76（p<.001），Cohen’s Kappa一致性係數為0.58（p<.001），顯示Sniffin’ sticks嗅覺辨識度測試在間隔1-2小時內具有良好的再測信度，而工具的最小可偵測變化值為3.4分。 結論：文獻中慢性腎臟病病人嗅覺辨識度異常的盛行率為48-92%，然本研究嗅覺辨識異常比率僅20％，可能是由於以高標準的認知功能評估排除輕度認知功能障礙者及受試者的高教育程度所致。Sniffin’ sticks嗅覺辨識測試在間隔1-2小時內具有良好的再測信度，顯示過去研究中經透析治療後嗅覺辨識度的改變非工具造成的測量誤差，最小可偵測變化值為3.4分，代表兩次分數的差距要超過3.4分，嗅覺改變才具臨床意義。 關鍵字：第五期腎臟病病人、嗅覺辨識、Sniffin’ sticks嗅味棒、再測信度

Background: The accumulation of uremic toxins damages the intrinsic antioxidant systems of the olfactory epithelium and olfactory bulb, leading to smell dysfunction in chronic renal disease patients. Previous studies reported that olfactory identification was improved after the hemodialysis episode, but there’s no research to prove the short term reliability of the tool, therefore we can’t be sure whether the improvement of olfactory identification is caused by the measurement errors. This study aim to investigate the olfactory identification in stage 5 renal disease patients by using Sniffin’ sticks, also to test the tool’s short term reliability and the minimal detectable change. Method: This longitudinal research was conducted in the Nephrology outpatient department of National Taiwan University Hospital, from May 1, 2020 to October 31, 2020. We recruited patients who was not only diagnosed with stage 5 renal disease (eGFR< 15ml/min/1.73m2) and had not yet undergone hemodialysis treatment but also got higher than 23.5 score in Montreal Cognitive Assessment. The olfactory identification were tested two times using Sniffin’ sticks within 1 to 2 hours, and we disrupted the sequence of the Sniffin’ sticks in the second test to prevent the learning effect. Result: 60 participants were included in this study, The mean age was 63.6± 14.1 years old, and 75% of our participants had high educational attainment (above high school).The prevalence rate of olfactory identification dysfunction was 20%, None of the baseline characteristics (age, gender, diabetes mellitus, estimated Glomerular Filtration Rate, BUN, creatinine) did related to the olfactory identification dysfunction. For olfactory identification test, Intraclass correlation value was 0.76 (p<0.001), and Cohen’s Kappa value was 0.58 (p<.001), The minimal detectable change value of Sniffin’ sticks was 3.4 . Conclusion: The prevalence of olfactory identification dysfunction in chronic renal disease patients was 48-92%, though the prevalence of olfactory identification dysfunction was only 20% in our study. It’s assumed that we excluded patients with mild cognition impairment by using high standard of cognition assessment, Also, our participants had high educational attainment. Sniffin’ sticks has a good test-retest reliability within 1-2hours, which we can be sure that the improvement of olfactory identification from previous studies wasn’t caused by the measurement error. Our minimal detectable change was 3.4, which mean that only over 3.4 of difference change in olfactory function has its clinical meaning. Keywords: End Stage Renal Disease, olfactory identification, Sniffin’ sticks, test-retest reliability