某醫學中心心房纖維顫動病患特性與其抗血栓劑處方型態之研究

Abstract

心房纖維顫動（atrial fibrillation，ICD-9 427.31）病患比心跳為竇性節律的病患有5 ~ 17倍的中風危險性。心房纖維顫動的主要治療目標之ㄧ是預防中風。從1989年至1993年有六個大型臨床實驗證實warfarin可以降低心房纖維顫動之中風危險性達62%，各大相關醫學會亦針對心房纖維顫動病患抗血栓劑的選用提出治療建議，然而實際臨床上的應用卻是使用率不足。 本研究納入民國89年1月1日至93年7月30日曾在台大醫院門診、急診或住院被診斷為心房纖維顫動病患。依照2001年ACC/AHA/ESC準則建議，將病患分為四個危險等級，進行以下分析:（一）病患基本資料，包含性別、年齡及同時併有的可能致病因子分佈；（二）病患所使用之抗血栓劑處方型態分析；（三）warfarin的使用與準則的一致性探討；（四）warfarin的使用與缺血性中風或暫時性腦部缺血發作的相關性；（五）warfarin的使用與大腦出血或胃腸道出血的相關性；（六）使用warfarin病患穩定期之INR值；及（七）warfarin開方的預測因子。 本研究共納入了5,448位病患，平均年齡67±14歲（0 ~ 92歲），男性與女性的比例為1.25 : 1。病患同時併有的可能致病因子，最常見的是高血壓（54.6%）、冠狀動脈疾病（40%）、心臟衰竭（26.6%）和糖尿病（21.6%）。依據ACC/AHA/ESC準則，各等級的病患分佈為最低危險因子9%，低危險因子10%，高危險因子69%，最高危險因子13%。 Warfarin的使用與缺血性中風或暫時性腦部缺血發作相關性分析，761位使用warfarin的心房纖維顫動病患，有32人的住院診斷裡有缺血性中風（ICD-9 434）或暫時性腦部缺血發作（ICD-9 435），在校正年齡、性別對缺血性中風或暫時性腦部缺血發作的影響後，所得的勝算比為0.33（95% CI，0.22-0.47）。基於本研究是以人而非人次做為分析，因此可能低估了實際缺血性中風或暫時性腦部缺血發作的發生率。在處方行為上，有169人是在發生缺血性中風或暫時性腦部缺血發作後才開始使用warfarin。 Warfarin的使用與大腦出血或胃腸道出血相關性分析，991位使用warfarin的心房纖維顫動病患，有35人的住院診斷裡有大腦出血（ICD-9 431）或胃腸道出血（ICD-9 578），在校正年齡、性別後之勝算比為0.51（0.36-0.73）。如單獨分析胃腸道出血，於校正年齡、性別後勝算比為0.52（0.35-0.88）。結果顯示出現使用warfarin反而比較不會發生胃腸道出血。這可能是因為引起胃腸道出血原因多，像是消化性潰瘍、壓力性潰瘍等，因此使用warfarin引起的胃腸道出血相對的少很多。然而在單獨分析大腦出血，校正後勝算比為1.77（0.43-7.30），則是warfarin有無使用並不影響其發生率。大腦出血或胃腸道出血的發生率與年齡成正比，且出血時的INR值以3.5（34%）居多。本研究之病患群所維持的INR值為1.6 ±0.5。在warfarin開方的預測因子分析，老年人（≧ 85歲）是warfarin開方的障礙，本研究老年人比起年輕人的warfarin開方勝算比為 0.32，此結果和其他的研究相近（勝算比0.25-0.60）。心臟衰竭和曾發生缺血性中風或暫時性腦部缺血發作，是比較會使用warfarin。 本研究提供了國人心房纖維顫動人口統計學的參考數據。使用warfarin引起的大腦出血或胃腸道出血，與病患年齡、INR值相關。老年人是使用warfarin的開方障礙，在本研究或國外文獻皆有同樣結果。然而此病患群體發生缺血性中風或暫時性腦部缺血發作的危險性最高，因此老年人使用warfarin，是需要醫療團隊的合作，除了醫師外，臨床藥師亦可扮演重要的角色。同時國人需要有自己的數據，以使得病患在使用warfarin時得到的益處大於風險。

Patients with atrial fibrillation (AF, ICD-9 427.31) have a 5 to 17 fold increased risk of stroke when compared with patients without AF. Stroke prevention has become the standard of care. Six large randomized trials of stroke prevention in AF published from 1989 to 1993 had consistenly shown the benefit of warfarin therapy, with a risk reduction of 62%. Despite consensus that warfarin is strongly indicated in most patients with atrial fibrillation, past studies demonstrate that anticoagulation in AF patients is inadequately used. This study included patients visited National Taiwan University Hospital (NTUH) with a diagnosis of AF, from 2000 Jan 1 to 2004 July 30. We used ACC/AHA/ESC guideline released in 2001 to stratify our patients into 4 levels. We analyzed (1) demographic characteristics, (2) prescribing patterns of antithrombotic agents, (3) warfarin prescribing adherence with practice guidelines, (4) the association between warfarin usage and ischemic stroke or TIA complications, (5) the association between warfarin usage and cerebral or GI bleeding complications, (6) INR levels of stable patients under warfarin treatment, and (7) evaluation of the predictor factors of warfarin prescription. In our study, we included 5,448 patients, the average age (± SD) was 67 ±14 years (range, 0-92 years). The prevalence was higher in men (1.25 : 1). The most frequent comorbidities in patients with AF were hypertension (54.6%), coronary artery disease (40%), heart failure (26.6%), and diabetes mellitus (21.6%). Based on ACC/AHA/ESC guideline, our patient distribution from lowest risk to highest risk were 9%, 10%, 69%, and 13% respectively. Among 761 patients with warfarin treatment for atrial fibrillation, 32 hospital admissions with the diagnosis of ischemic stroke (ICD-9 434) or TIA (ICD-9 435). After adjusted age and sex, the odds ratio was 0.33 (95% CI, 0.22-0.47). Since we use person and not episode in our analysis, we might underestimate the incidence of ischemic stroke or TIA. In prescribing behavior, 169 patients were prescribed with warfarin after ischemic stroke or TIA. Among 991 patients with warfarin treatment for atrial fibrillation, 35 hospital admissions with the diagnosis of cerebral hemorrhage (ICD-9 431) or GI bleeding (ICD-9 578), after adjusted age and sex, the odds ratio was 0.51 (95% CI, 0.36-0.73). When we analyzed GI bleeding alone, the adjusted odds ratio was 0.52 (95% CI, 0.36-0.88). Our findings are unsual in that warfarin usage is decrease the risk of GI bleeding. Several reasons may, in part, explain this finding. There were some confounding factors of GI bleeding, we didn’t exclude patients with peptic ulcers or stress ulcers. Compared with ulcers, warfarin-induced GI bleeding was relatively low. In the other way, we analyse cerebral hemorrhage alone, the adjusted odds ratio was 1.77 (95% CI, 0.43-7.30). Warfarin usage was lack association with cerebral hemorrhage. The incidence of cerebral hemorrhage or GI bleeding increased with age, and most patients had INR level of 3.5 (34%). The INR levels of stable patients under warfarin treatment was 1.6