- 學位論文
雙極性情感性疾病患者情緒穩定劑使用與相關副作用之研究
Studies of Utilization of Mood Stabilizers and Related Adverse Drug Reactions in Bipolar Patients
摘要
研究背景 躁鬱症用藥的選擇包括情緒穩定劑(lithium、valproate、carbamazepine)、抗精神病藥物與抗憂鬱劑等。過去的研究顯示，lithium可能造成甲狀腺功能低下，而carbamazepine則和SJS/TEN的發生有關。目前國內無躁鬱症患者相關的精神藥物流行病學研究報告，因此，針對躁鬱症患者進行研究，可提供為臨床用藥及精神醫療政策的重要參考。 研究目的 針對躁鬱症病患使用的精神科藥物進行藥物處方型態分析研究，且以甲狀腺功能低下症及嚴重皮膚病變為主題，探討情緒穩定劑的使用與藥物不良反應之相關性。 研究方法 以台灣全民健康保險資料庫之「精神疾病住院病患歸人檔(PIMC)」為研究材料，由PIMC資料庫中取得躁鬱症病患在1997-2004年之就醫資料，進行長期回顧性世代研究。分析情緒穩定劑、非典型抗精神病藥物、典型抗精神病藥物與抗憂鬱劑四大類用藥，以及電痙攣療法的使用趨勢。採用病例對照研究設計，首先依納入條件篩選出病例組病人，並依性別、年齡為條件，以1:4比例隨機選出對照組病人，進行描述性統計分析，並利用邏輯式回歸分析控制其他危險因子計算勝算比(odds ratio)，探討甲狀腺功能低下症的發生與情緒穩定劑使用之相關性，及嚴重皮膚病變的發生與carbamazepine及其他相關藥物使用之相關性。 研究結果 由PIMC資料庫，共擷取出12,424位躁鬱症病患，其平均年齡為40.2歲，男女比例相近。自1997至2004年間，不論門診或住院期間之處方，非典型抗精神病藥物的使用呈顯著上升的趨勢(β=0.42,p<0.0001)、典型抗精神病藥物呈顯著下降的趨勢(β=-0.23,p<0.0001)。以單一情緒穩定劑分析，valproate的使用有逐年上升的顯著趨勢(β=0.25,p<0.0001)，而lithium (β=-0.091, p<0.0001)與carbamazepine (β=-0.096, p<0.0001)則相對有顯著下降。由12,424位躁鬱症患者中共篩選出557位甲狀腺功能低下症病例組個案與2,228位對照組個案，其男女比例約為1:3，平均年齡為41.3歲。經由多變項分析後發現，lithium的使用相較於valproate與carbamazepine，對甲狀腺功能低下症並未有顯著危險性；用過lithium與valproate及三種情緒穩定劑皆用過者與症狀的發生有顯著相關性，且用過愈多種情緒穩定劑的患者，其發生甲狀腺功能低下症的危險性愈高。由12,424位躁鬱症患者中初步篩選出74位嚴重皮膚病變個案，後移除兩位在index date前60天內無藥物處方者，共得72位病例組個案，男女比例約1:1.06，平均年齡為41.3歲，經配對後得288位對照組個案。於多變項分析後發現，carbamazepine與valproate的使用與事件有顯著相關性，將使用人數較少的抗癲癇藥物phenytoin、phenobarbital與lamotrigine視為同一群體分析，亦呈現顯著相關性。 結論 台灣躁鬱症患者的用藥型態於1997至2004年間有顯著改變，其改變趨勢應與國際上躁鬱症用藥準則的建議內容改變有關。曾使用過兩種以上情緒穩定劑的病患，發生甲狀腺功能低下症的危險性有顯著的增加，臨床用藥上應予以注意。躁鬱症患者發生SJS/TEN的相關藥物中，以carbamazepine的危險性最高，而其他抗癲癇劑的危險性亦不容忽視，病患在使用抗癲癇藥物時應提高對皮膚不良反應的警覺性。
並列摘要
Background The mood stabilizers for bipolar disorder include lithium, valproate and carbamazepine. According to previous studies, lithium and carbamazepine were considered to be associated with the development of hypothyroidism and SJS/TEN, respectively. The pharmacoepidemiology data for bipolar patients is important for clinical practice and medical policy. Objective To analyze the prescription pattern of bipolar disorders from 1997 to 2004. To evaluate the adverse drug reactions, i.e. hypothyroidism and SJS/TEN, of mood stabilizers in bipolar patients. Methods The Psychiatric Inpatient Medical Claim dataset (PIMC) in Taiwan was used in this study. Medical claim data between 1997 to 2004 of the bipolar patients diagnosed during 1996-2001 were retrieved. The utilization patterns of mood stabilizers, antipsychotics, antidepressants and electroconvulsive therapy were calculated and analyzed. The matched case-control design with one case to four controls was used for the ADR studies. Bipolar patients who fit the inclusion criteria of case group were screened first. The control group of the same age range and sex to the cases was retrieved from patients other than the case group. Conditional logistic regression model using to calculate the adjusted odds ratio was applied to analyze the relationship between hypothyrodism or severe skin reactions and the drugs used by the bipolar patients. Results A total of 12,424 patients of bipolar disorder who had a mean age of 40.2 year old and equal gender ratio were included. From 1997-2004, the use of atypical antipsychotics (β=0.42, p<0.0001) increased significantly with a decrease in typical antipsychotics (β= - 0.23, p<0.0001). For mood stabilizers, the prescriptions of valproate (β=0.25, p<0.0001) increased significantly along with the significant decreases of lithium (β=-0.091, p<0.0001) and carbamazepine (β=-0.096, p<0.0001). A total of 557 bipolar patients with hypothyroidism were identified and 2,228 controls were selected. The mean age of these patients was 41.3 year old with a ratio of man to women 1:3. After adjustment, lithium did not exhibit higher risk in hypothyroidism as compared to carbamazepine and valproate. Patients who have ever used both lithium and valproate or who have ever used all the three mood stabilizers have significant risk of developing hypothyroidism. If considered the number of mood stabilizers used as a continuous factor, it was found that the more mood stabilizers used the higher risk in developing hypothyroidism (OR 1.34, 95% CI 1.21-1.49). There were 74 cases with severe skin reactions and two of them did not have any prescription within the 60 days before the index date. Finally, 72 cases and 288 controls were identified from the 12,424 bipolar patients. The number of man and women were about the same with a mean age of 41.3 years old. After multivariate analysis, carbamazepine and valproate were found significantly associated with severe skin reactions. The other anticonvulsants also had significant association as calculated as a group of drugs. Among the suspicious drugs, carbamazepine showed the highest risk in developing the severe skin reactions (OR 4.02, 95% CI 1.97-8.27). Conclusions The prescription pattern of bipolar disorder during 1997 to 2004 has a significant change and this change seems to be consistent with recent guidelines. Using more than two mood stabilizers may lead to a higher risk of hypothyroidism. Closely monitor thyroid function for these patients may be necessary. Carbamazepine has the highest risk of causing severe skin reactions among all suspicious drugs in bipolar patients. The association of other anticonvulsants also could not be ignored. It is important to be aware of shin reactions during anticonvulsants treatment.