缺血性腦中風病人昏迷狀態與住院醫療利用及照護結果之探討

研究背景：腦中風為全球第二大死因和導致失能第三大原因，而罹患缺血性腦中風對家庭和社會產生龐大疾病負擔，台灣腦血管疾病在2017‒2020年期間醫療費用位居前7名。過去研究指出格拉斯哥昏迷指數是最常用於評估病人昏迷狀態，並用於探討腦中風病人住院費用和死亡率，因此本研究欲使用格拉斯哥昏迷指數(Glasgow coma scale, GCS)檢視缺血性腦中風病人到院時昏迷狀態與其住院醫療資源利用和照護結果是否存在差異。目的：本研究主要目的為：(1)探討病人不同昏迷狀態與住院醫療利用之相關性；(2)探討病人不同昏迷狀態與住院照護結果之相關性。材料與方法：本研究採用回溯型世代研究設計，資料來源取自高雄地區某區域教學醫院缺血性腦中風病人住院資料檔，使用國際疾病分類第十版臨床修訂版診斷代碼為I63，撈取於2019年1月1日至2021年12月31日期間年齡≥20歲、罹患缺血性腦中風病人住院醫療利用及照護結果，而最終樣本數納入1,252筆。變項包括昏迷狀態(使用入院時格拉斯哥昏迷指數(GCS)進行評估)、入院年齡、性別、身體體質指數、吸菸、飲酒、高血壓、糖尿病、心血管疾病、高血脂、心臟疾病、使用抗血栓藥物、使用血栓溶解劑，以及住院醫療利用和照護結果。使用廣義線性模式分析缺血性腦中風病人昏迷狀態和相關因素分別對住院天數和住院總費用之影響；二元羅吉斯迴歸分析缺血性腦中風病人昏迷狀態和相關因素分別對是否接受復健治療、併發肺炎、尿道感染、上消化道出血、癲癇、出院時生活功能依賴及死亡之影響。結果：缺血性腦中風病人昏迷狀態以輕度最多(83.3%)，其次為中度(11.7%)，而重度最少(5%)；年長者(65%)、男性(64.2%)和身體質量指數標準者(18.5‒24公斤/公尺2，35.8%)比例較高；有飲酒史(78.5%)或高血壓者(78.5%)最多，其次為血脂異常(61.7%)、糖尿病(41%)、有吸菸史(36.3%)、腦中風病史(24.3%)。另外，病人使用抗血栓藥物者較多(98.4%)，其次為使用血栓溶解劑者(7.2%)。病人併發肺炎比例最高(11.9%)，其次為尿道感染(9.9%)、上消化道出血(4.2%)；有10.1%接受復健治療，出院時生活功能需依賴他人者約46.9%，平均住院天數為11.9天(中位數：9天)；每人平均住院費用為84,155.4點值(中位數：48,771.5點值)，而死亡率為3.5%。昏迷狀態為重度者以老年(84.1%)居多，而輕度者以壯年(32.4%)居多，且達統計顯著差異(p=0.001)。重度者以男性(50.8%)居多，而輕度者以女性(66.9%)居多，且達統計顯著差異(p<0.001)。重度者以肥胖(>27公斤/公尺2，27%)居多，輕度者以過重(<18.5公斤/公尺2，32%)居多，且達統計顯著差異(p=0.007)。重度者(16天)住院天數中位數較高於中度(14天)和輕度者(8天)，且達統計顯著差異(p<0.001)；重度者(137,839點值)住院費用中位數較高於中度(96,279點值)和輕度(42,691點值)，且達統計顯著差異(p<0.001)。再者，廣義線性模式分析結果顯示重度者(OR=0.84，95%CI: 1.09‒65.39，p=0.041)接受復健治療機率較低於輕度者；中度(B=4.79，95% CI: 2.96‒6.61，p<0.001)與重度者(B=7.01，95% CI: 3.75‒10.27，p<0.001)住院天數較長於輕度者、住院費用較高(B=55,613.31，95%CI: 32119.09‒79107.52，p<0.001；B=99,052.52，95%CI: 58,759.89‒139,345.16，p<0.001)。二元羅吉斯迴歸分析結果顯示中度與重度者併發肺炎(OR=4.27，95%CI: 2.71‒6.73，p<0.001；OR=6.13，95%CI: 3.37‒11.15，p<0.001)、併發泌尿道感染(OR=2.77，95%CI: 1.69‒4.56，p<0.001；OR=2.28，95%CI: 1.14‒4.58，p=0.02)、併發上消化道感染(OR=2.04，95%CI: 0.99‒4.23，p=0.055；OR=2.74，95%CI: 1.09‒6.84，p=0.032)、併發癲癇(OR=2，95%CI: 0.36‒11.28，p=0.431；OR=8.36，95%CI: 1.78‒39.28，p=0.007)、生活功能依賴(OR=4.81，95%CI: 3.06‒7.58，p<0.001；OR=7.15，95%CI: 3.20‒15.96，p<0.001)，以及死亡(OR=5.62，95%CI: 2.5‒12.62，p<0.001；OR=17.15，95%CI: 7.34‒40.1，p<0.001)機率皆較高。結論：昏迷狀態越嚴重的缺血性腦中風病人住院天數和住院費用明顯較多於輕度者，而出院後持續接受復健比例明顯較低，且其併發肺炎、尿道感染、上消化道出血、癲癇、出院時生活功能依賴以及死亡機率亦明顯較高。本研究建議衛生主管機關和醫院醫療照護團隊可針對昏迷狀態較嚴重的缺血性腦中風病人提供即時監測與感染控制措施，以確保住院照護品質和延緩病況惡化結果。

關鍵字

缺血性腦中風；昏迷狀態；住院醫療利用；住院照護結果

並列摘要

Background: Stroke is the second leading cause of death globally and the third leading cause of disability, with ischemic stroke accounting for approximately 87% of cases. The burden of ischemic stroke on families and society is substantial. In Taiwan, from 2017 to 2020, medical expenses for cerebrovascular diseases consistently ranked among the top seven. Previous studies have indicated that the Glasgow Coma Scale (GCS) is commonly used to assess the conscious status of patients and is utilized to explore hospital costs and mortality rates in stroke patients. Therefore, this study intends to use the GCS to examine whether there are differences in hospitalization utilization and care outcomes among ischemic stroke patients based on their comatose states upon hospital admission. Objectives: The primary purposes of this study are: (1) to explore the relationship between different comatose states of patients and hospital medical resource utilization. (2) To explore the relationship between different comatose states of patients and care outcomes. Materials and Methods: This research adopted a retrospective cohort study design. Data were obtained from a regional teaching hospital in Kaohsiung. The patient data pertained to individuals hospitalized for ischemic stroke, identified using the ICD-10-CM code I63. Data collection spanned from January 1, 2019, to December 31, 2021, including patients aged ≥20 years. The final sample comprised 1,252 cases. Variables included conscious status (assessed using the GCS at admission), age at admission, gender, body mass index (BMI), history of smoking or alcohol use, hypertension, diabetes mellitus, cardiovascular diseases, hyperlipidemia, heart diseases, use of antithrombotic drugs, thrombolytics, and hospitalization utilization and care outcomes. Binary logistic regression was performed to analyze the impact of conscious status and related factors on whether patients received rehabilitation therapy, had pneumonia, urinary tract infection, upper gastrointestinal bleeding, seizures, dependency in activities of daily living at discharge, and death. Generalized linear models were employed to examine the impact on hospital length of stay and total hospital costs. Results: Among ischemic stroke patients, those with mild coma were the most common (83.3%), followed by moderate (11.7%) and severe coma (5%). The majority were elderly (65%), male (64.2%), and had a standard BMI (35.8%). A history of alcohol use (78.5%) or hypertension (78.5%) was most prevalent, followed by dyslipidemia (61.7%), diabetes mellitus (41%), history of smoking (36.3%), and stroke history (24.3%). The majority used antithrombotic drugs (98.4%), while 7.2% used thrombolytics. Pneumonia was the most common complication (11.9%), followed by urinary tract infections (9.9%) and upper gastrointestinal bleeding (4.2%). Rehabilitation therapy was received by 10.1% of patients, and 46.9% were dependent on others for daily activities at discharge. The average length of hospital stay (LOS) was 11.9 days (median: 9 days), and the average hospitalization cost was 84,155.4 points (median: 48,771.5 points). The mortality rate was 3.5%. Patients with severe coma were mostly elderly (84.1%), while those with mild coma were mostly middle-aged (32.4%), with a statistically significant difference (p=0.001). Severe coma patients were predominantly male (50.8%), while those with mild coma were predominantly female (66.9%), with a statistically significant difference (p<0.001). Besides those with a standard BMI, severe coma patients were more likely to be obese (27%), while mild coma patients were more likely to be overweight (32%), with a statistically significant difference (p=0.007). The median LOS was more extended for severe coma patients (16 days) compared to moderate (14 days) and mild coma patients (8 days), with a statistically significant difference (p<0.001). The median hospitalization cost was higher for severe coma patients (137,839 points) compared to moderate (96,279 points) and mild coma patients (42,691 points), with a statistically significant difference (p<0.001). Further analyses showed that patients with severe coma had a lower likelihood of receiving rehabilitation therapy compared to those with mild coma (OR=0.84, 95%CI: 1.09‒65.39, p=0.041). Patients with moderate (B=4.79, 95%CI: 2.96‒6.61, p<0.001) and severe coma (B=7.01, 95%CI: 3.75‒10.27, p<0.001) had more extended hospital stays and higher hospitalization costs (B=55,613.31, 95%CI: 32119.09‒79107.52, p<0.001; B=99,052.52, 95%CI: 58,759.89‒139,345.16, p<0.001), a higher likelihood of pneumonia (OR=4.27, 95%CI: 2.71‒6.73, p<0.001; OR=6.13, 95%CI: 3.37‒11.15, p<0.001), urinary tract infections (OR=2.77, 95%CI: 1.69‒4.56, p<0.001; OR=2.28, 95%CI: 1.14‒4.58, p=0.02), upper respiratory infections (OR=2.04, 95%CI: 0.99‒4.23, p=0.055; OR=2.74, 95%CI: 1.09‒6.84, p=0.032), seizures (OR=2, 95%CI: 0.36‒11.28, p=0.431; OR=8.36, 95%CI: 1.78‒39.28, p=0.007), dependency in daily activities (OR=4.81, 95%CI: 3.06‒7.58, p<0.001; OR=7.15, 95%CI: 3.20‒15.96, p<0.001), and mortality (OR=5.62, 95%CI: 2.5‒12.62, p<0.001; OR=17.15, 95%CI: 7.34‒40.1, p<0.001) compared to those with mild coma. Conclusions: Patients with more severe coma have a lower rate of continuing rehabilitation after discharge, and their hospital stays and costs are significantly higher than those of patients with mild coma. Additionally, they have a significantly higher likelihood of developing pneumonia, urinary tract infections, upper gastrointestinal bleeding, seizures, dependence on others for daily activities at discharge, and mortality. This research recommends that health authorities provide closer monitoring and infection control measures for patients with severe coma and strengthen education and psychological support for their families to improve treatment outcomes and quality of life for ischemic stroke patients.