透析患者需要投與紅血球生成素，以維持適當血比容。國內為鼓勵腹膜透析，將紅血球生成素給付規範之血比容限制放寬，原先血比容超過30%不予給付，在民國95年11月1日將血比容上限放寬至36%，超過36%方不予給付；唯每個月紅血球生成素的使用劑量仍限制在20,000單位(Eprex、Recormon)或100mcg (Aranesp)以下。國內腹膜透析患者之血比容在放寬前約為28-29%，或許因紅血球生成素給付劑量的限制，在血比容給付上限放寬後，亦僅上升至30-31%。
關於血比容的目標值，近年來幾個大型試驗如Normal Hematocrit Study (Besarb A, et al. 1998)、CHOIR Study (Singh AK, et al. 2006)與TREAT Study (Pfeffer MA, et al. 2010)結果顯示，血比容過高(>37%)會伴隨較多的心血管事件。美國FDA於2011年6月建議透析患者血比容維持在30-33%。換言之，低血比容(30-33%)較高血比容(>37%)為安全。然國內腹膜透析患者之血比容在給付條件放寬前僅28-29%，過於偏低，在放寬後方增至30-31%。放寬後維持血比容在30-31%，是否較放寬前維持在28-29%為安全，即是否可減少之心血管事件風險，有賴國人資料加以分析。
本研究為利用「全民健康保險資料庫」，比較健保紅血球生成素給付條件放寬前後，對腹膜透析患者心血管事件風險之影響。紅血球生成素放寬的健保給付規範於95年11月1日施行，故以95年11月1日為切點，於放寬前(Time 1)與放寬後(Time 2)分別篩選出初次腹膜透析患者，追蹤心血管事件(包括心肌梗塞、中風、因心衰竭住院、死亡)發生率。本研究採取兩種方式定義初次腹膜透析患者，寬鬆法與嚴謹法。寬鬆法不限定開始透析的第一至三個月的透析模式，僅限定開始透析的第四個月須為腹膜透析。嚴謹法除開始透析的第四個月須為腹膜透析外，尚須有連續三個月為腹膜透析。
研究結果顯示，以寬鬆法所得Time 1有1759人，Time 2有2981人；嚴謹法Time 1有1733人，Time 2有2940人。Time 2有較多高血壓與糖尿病之患者。紅血球生成素每個月使用劑量之中位數，在Time 2為13,105U，Time 1為10,769U，兩者差異達統計上顯著(p<0.0001)。以寬鬆法所得之Time 1與Time 2初次腹膜透析患者，經校正年齡、性別、Charlson index、糖尿病、高血壓、冠狀動脈繞道手術病史、心衰竭病史等變項後，針對心血管事件綜合指標以Cox proportional hazard model分析，Time 2相對於Time 1之hazard ratio為0.823，95% 信賴區間為0.711，0.954，達統計上顯著降低(p=0.0096)。就個別心血管事件，包括「心肌梗塞」、「中風」、「因心衰竭住院」、「死亡」，則Time 2相對於Time 1之hazard ratio均小於1，唯未達統計上顯著。以嚴謹法所得之結果與寬鬆法相同。進一步檢視各次族群結果，發現無糖尿病的族群，Time 1與Time 2心血管事件無顯著差異。而糖尿病族群在Time 2相對於Time 1心血管事件綜合指標風險明顯降低(hazard ratio 0.737, 95%信賴區間0.613，0.886，p=0.0011)；就個別心血管事件，「中風」在Time 2相對於Time 1風險達顯著降低(hazard ratio 0.591, 95%信賴區間0.396，0.883，p=0.0102)，「因心衰竭住院」在Time 2相對於Time 1風險較低，差異接近統計上顯著(hazard ratio 0.785, 95%信賴區間0.614，1.004，p=0.0535)。以嚴謹法所得之結果與寬鬆法相同。
本研究結果可支持在限定紅血球生成素最高使用劑量下，將血比容維持在30-31%較28-29%為安全，可降低心血管事件風險。對於是否需將健保給付規範中血比容上限調降，則有待進一步討論。此外，是否紅血球生成素最高劑量限制，對於糖尿病患者更有利，則有待後續試驗加以驗證。

Background: Erythropoietins are necessary for dialysis patients to maintain adequate hematocrit(Hct). In order to promote peritoneal dialysis, the erythropoietin payment criteria loosened in Nov 2006. Before Nov 2006, erythropoietin was not covered by National Health Insurance (NHI) if the Hct was above 30%. After Nov 2006, the erythropoietin payment criteria loosen and erythropoietin can be covered by NHI if the Hct is below 36%. However, the maximum monthly erythropoietin dosage remains the same (Eprex/Recormon 20,000U, Aranesp 100mcg). After widening the benefit package, the Hct of prevalent peritoneal dialysis patients increased from 28-29% to 30-31%. From the results of Normal Hematocrit study, CHOIR study and TREAT study, we find that administration of erythropoietins and pushing Hct to more than 37% is associated with adverse cardiovascular outcome. However, what is the optimal Hct target is still an issue of debate. We need local data to find out if it is better to maintain Hct between 30-31% rather than 28-29%.
Objective: The main purpose of this study is to analyze the impact of loosening erythropoietin payment criteria on cardiovascular outcome of incident peritoneal dialysis patients.
Methods: We selected incident peritoneal dialysis patients from the NHI beneficiaries claim data from 2003 to 2010. We defined the period before loosening the erythropoietin payment criteria as Time 1 (from May 2003 to Oct 2006), the period after loosening the payment criteria as Time 2 (from May 2007 to Oct 2010). Incident peritoneal dialysis patients were followed for cardiovascular events including myocardial infarction, stroke, hospitalization due to heart failure and death.
Findings: There are 1759 incident peritoneal dialysis patients in Time 1 and 2981 patients in Time 2. More patients in Time 2 have hypertension and diabetes mellitus (DM). The median monthly erythropoietin dosage is significantly higher in Time 2 (Time 1: 10,769U; Time 2: 13,105U). For the composite cardiovascular endpoint, the risk in Time 2 is significantly lower (HR 0.823, 95% confidence interval 0.711, 0.954, p=0.0096) after adjusting age, sex, Charlson index, DM, hypertension, history of coronary artery bypass graft and congestive heart failure. For each cardiovascular endpoint, the risk reduction in Time 2 does not reach statistical significance. With regard to subgroup analysis, for patients without DM, no significant difference in cardiovascular risk is observed between Time 1 and Time 2. However, for patients with DM, the risk of composite cardiovascular endpoint is significantly lower in Time 2 (HR 0.737, 95% confidence interval 0.613, 0.886, p=0.0011). In addition, the risk of stroke is also significantly lower in Time 2 (HR 0.591, 95% confidence interval 0.396, 0.883, p=0.0102). The risk of hospitalization due to heart failure is lower in Time 2, but does not reach statistical significance (HR 0.785, 95% confidence interval 0.614, 1.004, p=0.0535). The above data can support that under maximum monthly erythropoietin dose limit, less cardiovascular risk is observed when Hct is maintained within 30-31% compared to 28-29%. However, it needs further discussion to see if refine the erythropoietin payment criteria is necessary. Current payment criteria can leave more room for clinical practice. Besides, further studies are needed to prove that whether DM patients can have less cardiovascular events with Hct 30-31% under maximum monthly erythropoietin dose limit

口試委員會審定書 i
誌謝 ii
中文摘要 iii
Abstract v
目錄 viii
表目錄 ix
圖目錄 x
第一章緒論 1
第一節研究背景與動機 1
第二節研究目的 3
第二章文獻探討 4
第一節慢性腎臟病對紅血球生成素之影響 4
第二節透析患者血比容(Hct)的臨床意義 5
第三節國內紅血球生成素與健保給付 9
第三章研究方法 13
第一節研究設計與研究樣本 13
第二節研究假說 16
第三節研究變項 16
第四節統計分析方法 17
第四章研究結果 19
第一節樣本基本特質之描述性統計 19
第二節對心血管事件綜合指標與個別心血管事件之風險比 22
第三節對心血管事件綜合指標與個別心血管事件之KM分析 24
第四節糖尿病與非糖尿病次群體之比較 27
第五章討論 30
第一節研究結果討論 30
第二節研究限制 35
參考文獻 36

中文部份

台灣血液透析診療指引。台灣腎臟醫學會編著，民國93年12月30日。
健保醫字第0900015412號，全民健康保險藥品給付規定修正規定，造血功能治療藥物-紅血球生成素，90年6月21日。
健保審字第0930006236號，全民健康保險藥品給付規定修正規定，4.1.造血功能治療藥物，93年4月15日。
健保審字第 0930021060 號，修正「全民健康保險藥品給付規定」：代謝及營養劑，民國 93 年 11 月 15 日。
健保藥字第0950070568號，全民健康保險藥品給付規定修正規定，4.1.造血功能治療藥物，95年9月22日。
楊五常，透析服務支付政策之評估，行政院衛生署101年度委託研究計畫研究成果期末報告，101年12月。
96年第四季門診透析總額品質報告，指標5.10 (94Q1 96Q4)；http://www.nhi.gov.tw/webdata/webdata.aspx?menu=17&menu_id=1027&WD_ID=1035&webdata_id=818
99年第三季門診透析總額品質報告，指標5.12 (97Q1 to 99Q3)http://www.nhi.gov.tw/webdata/webdata.aspx?menu=17&menu_id=1027&WD_ID=1035&webdata_id=818
謝其政 (2011)。台灣末期腎臟病患血液透析與腹膜透析成本效果分析。國立台灣大學公共衛生學院健康政策與管理研究所博士論文。
蔡雅馨 (2011)。門診透析總額制度對末期腎臟病患醫療利用與處方藥品之影響。國立台灣大學公共衛生學院健康政策與管理研究所碩士論文。
張宏瑋 (2011)。台灣2020年末期腎臟病透析治療發生人數與盛行人數預測。國立台灣大學公共衛生學院健康政策與管理研究所碩士論文。
英文部份

AranespUS FDA label, 17 Dec 2007 version.
AranespUS FDA label, 24June 2011 version.
Besarab A, Bolton WK, Browne JK, Egrie JC, et al.(1998). The effect of normal as compared with low hematocrit values in patients with cardiac disease who are receiving hemodialysis and epoetin. The New England Journal of Medicine, 339:584-90.
Brenner & Rector’s The Kidney. 2012
KDOQI Clinical Practice Guideline and Clinical Practice Recommendations for Anemia in Chronic Kidney Disease:2007 Update of Hemoglobin Target. (2007). American Journal of Kidney Disease, 50(3):477-512.
Epogen US FDA label, 17 Dec 2007 version.
Epogen US FDA label, 24 June 2011 version.
Eschbach JW, Abdulhadi MH, Browne JK, et al.(1989) Recombinant human erythropoietin in anemic patients with end-stage renal disease: results of a phase III multicenter clinical trial. Ann Intern Med, 111:992-1000.
European Medical Agency (EMA). Public Statement (2007)Epoetins and the risk of tumor growth progression and thromboembolic events in cancer patients and cardiovascular risks in patients with chronic kidney disease.Doc.Ref. EMEA/496188/2007
Evans RW, Rader B, Manninen DL (1990) Cooperative Multicenter EPO Clinical Trial Group. The quality of life of hemodialysis recipients treated with recombinant human erythropoietin. JAMA, 263:825-830.
Fishbane S, Besarb A (2007) Mechanism of increased mortality risk with erythropoietn treatment to higher hemoglobin targets. Clin J Am Soc Nephro, 2:1274-1282
Kaufman JS (2011) Relationship of erythropoiesis-stimulating agent dose and reponsiveness and adverse outcomes in CKD. Am J Kidney Dis, 57(5):661-663
Mircera US FDA label, 14 Nov 2007 version.
Mircera EMA label, 19 June 2012 version.
Pfeffer MA, Burdmann EA, Chen CY, et al. (2010). A trial of darbepoetin alfa in type 2 diabetes and chronic kidney disease. The New England Journal of Medicine, 361:2019-32.
Procrit US FDA label, 17 Dec 2007 version.
Procrit US FDA label, 24 June 2011 version.
Schrier RW.: Diseases of the Kidney and Urinary Tract. Lippincott Williams & Xilkins. 2001.
Singh AK, Szczech L, Tang KL, et al. (2006) Correction of anemia with epoetin alfa in chronic kidney disease.The New England Journal of Medicine, 355:2085-98.
Singh AK, Himmelfarb J, Szczech LA (2009) Resolved: Targeting a higher hemoglobin is associated with greater risk in patients with CKD anemia. J Am Soc Nephrol, 20:1436-1443
Singh AK (2010) Dose TREAT give the boot to ESAs in the treatment of CKD anemia? J Am Soc Nephrol,1-13 21:
Solomon SD, Uno H, Lewis EF, et al. (2010) Erythropoietin Response and outcomes in kidney disease and type 2 diabetes. N Engl J Med, 363(12):1146-1155.
Szczech L, Barnhart HX, Inrig JK, Reddan DN, et al. (2008) Secondary analysis of the CHOIR trial epoetin-α dose and achieved hemoglobin outcomes. Kidney International, 74:791-798
Macdougall IC, Lewis NP, Saunders MJ, et al. (1990)Long-term cardiorespiratory effects of amelioration of renal anemia by erythropoietin. Lancet, 335:489-493[Erratum, Lancet 1990; 335:614]
US Food and Drug Administration (FDA) Safety Announcement (2011) FDA Drug Safty Communication: Modified dosing recommendations to improve the safe use of Erythropoiesis-Stimulating Agenst (ESAs) in chronic kidney disease. 24 June.
U.S. Renal Data System. (2012). USRDS 2012 annual data report: Atlas of end-stage renal disease in the United States.
World Health Organization (2011) Hemoglobin concnetrations for the diagnosis of anemia and assessment of severity. WHO/NMH/NHD/MNM/11.1