情緒障礙患者的晝夜節律及睡眠參數之研究

Abstract

情緒障礙病患常常伴隨著晝夜節律紊亂，主要表現為睡覺和活動規模的改變。本篇研究主要比較在情緒疾患病人（躁鬱症、重度躁鬱症）和健康組之間畫夜節律和睡眠的參數，並且利用腕動計 (Actigraphy)測量的參數來辨別疾病的診斷。此外，我們嘗試解釋在情緒疾患病人之中憂鬱症狀嚴重程度與主客觀睡眠參數和活動節律參數的關係。 本篇研究納入來自精神門診的18個躁鬱症和32重度憂鬱症病人。25位健康組則是在社區招募沒有心理疾病的健康成人。利用匹茲堡睡眠品質量表測量主觀睡眠品質；貝氏憂鬱量表則用來測量憂鬱症狀嚴重程度。活動節律參數則是利用腕動計測量。腕動計會記錄參加者的連續七天的活動次數 （每個參加者共有10080筆採樣數據）。利用餘弦分析法（包括中值、振幅和峰值位相值）、無母數分析法（跨日穩定度、日內變異度和相對振幅）、平均活動量及平均久坐(小於58次／分鐘)時間來描述畫夜節律的表現。我們同時通過腕動計取得客觀睡眠參數和睡眠日誌取得主觀睡眠參數，其中睡眠參數則包括睡眠遲滯期、睡眠效率、睡後醒來的時間, 總臥床時間和總睡眠時間。在較正年齡及性別後，利用多元邏輯迴歸斯分辦健康組和病例組。。另外，同樣較正年齡及性別後，利用多元線性迴歸來解釋情緒疾患病人之中憂鬱症狀嚴重程度與主觀睡眠品質、主客觀睡眠參數和活動節律參數的關係。 結果顯示，在睡眠的面向中相對健康組的人，情緒疾患病人相比健康組的人有顯著更高的匹茲堡睡眠品質量表分數(p<0.001)，更長的總臥床時間(p=0.001~0.002)、睡眠遲滯期(p=0.021)和總睡眠時間(p=0.005)。另一方面，在活動節律面向結果只顯示憂鬱症的病人相對健康組的人有顯著更長的久坐時間(p=0.003)。早上能量感覺在三組之間存在顯著差異（p=0.038）。在多分類邏輯迴歸結果顯示，儘管模型只包括腕動計測量參數，卻依然足以分辦是否為病患(AUC值：0.881)，但並不能分辦疾病的類型 。同時在同一個結果發現，相對健康組而言，情緒障礙病人有顯著更高的活動振幅和更長的久坐時間。在利用多元線性迴歸探討憂慮嚴重程度與睡眠和活動節律的關係結果顯示，在較正年齡和性別後，主觀睡眠品質相比睡眠參數可以解釋更多的變異。更進一步，結果顯示在情緒疾患病人當中，活動節律的變化相比睡眠變化更能解釋憂鬱嚴重程度。 最後總結，在利用多元邏輯斯迴歸分辨健康組及臨床上情緒疾病的診斷時，腕動計測量的參數，由其在振幅和久坐時間的變項可以幫助有效地判別是否有病，並且發現加入更多的面向例如能量感覺、睡眠干擾，可以使模式有更好的判別力。除些之外，我們發現在情緒障礙病患中，異常的畫夜效應主要表現在主觀睡眠品質和活動節律的相位上。

Patients with mood disorders often suffered from circadian disturbances, including sleep complaints and changed activity patterns. The current study compared circadian rest-activity rhythms, and sleep parameters between patients with bipolar disorder (BP) or major depressive disorder (MDD), and healthy controls using subjective and actigraphic measurements. In addition, we examined the correlations between depressive symptoms severity and sleep (objectively and subjectively) and activity parameters among mood disorder patients. We enrolled outpatients of BP (N=18) and MDD (N=32) from psychiatric clinics. Healthy controls (N=25) were recruited from the community without a history of mental illness. Pittsburgh sleep quality index (PSQI) was used to assess sleep quality. The depressive symptoms severity was assessed by the Beck Depression Inventory (BDI). The circadian rest-activity rhythm was collected by actigraphy, including wake-sleep cycle and physical activity counts in 7 consecutive days (each sample had 10080 data). Cosinor analysis (MESOR, amplitude and acrophase), nonparametric analysis (interdaily stability and intradaily variability) and the average of activity and the time of sedentary (< 58counts/min) were processed to describe rest-activity rhythm. We assessed sleep parameters (sleep latency, sleep efficiency, wake after sleep onset, total time in bed and total sleep time) using both actigraphy and sleep diary. Multinomial logistic regression models were used to distinguish the diagnosis groups after adjusted for age and gender. Multiple linear regression models were used to examine the relationships between depressive symptoms severity and sleep quality, objectively and subjectively assessed sleep and activity parameters. We found that patient groups had significantly higher PSQI global score (p<0.001), longer total time in bed (p=0.001~0.002), sleep latency (p=0.021), and total sleep times (P=0.005) than controls. Among the activity parameters, MDD group has significantly longer sedentary time in daytime than controls (p=0.003). Morning feeling was different (P=0.038) between groups. We found that actigraphic measurement in logistic regression analysis can distinguish affective status but not at diagnostic status. Comparing with controls, patient groups exhibited higher amplitude and longer daytime sedentary. For the relationship between depressive severity and sleep parameters, we found that subjective sleep quality explained more variation than objective sleep parameters in the regression models after taken activity variables into account. Among patients with affective disorders, we found that activity rhythms influenced the levels of depressive symptoms more predominantly then sleep variables. In conclusion, actigraphic parameters, including amplitude and sedentary times can distinguish affective disorders from health control, including more domains, such as energy feeling and sleep disturbances in the regression models could improve the prediction. We found that subjective poor sleep quality and the phase of activity rhythm are the primary expressions of circadian disturbances in mood disorder patients.