2018-08-012024-05-17https://scholars.lib.ntu.edu.tw/handle/123456789/686115摘要：常見的神經重症疾病主要為創傷性腦損傷、缺血性中風、及出血性中風等。因機車事故多，臺灣的頭部外傷比其它國家嚴重，死亡率亦高。此外，台灣每年約有一萬七千人會因為腦中風而導致日常生活失能，是成人殘障的第一要因，也是健保資源耗用前三名的疾病。神經重症的病患若需手術治療，在手術後，皆需使用呼吸器在加護病房密切治療照護。謹慎的處理呼吸器的使用策略，可避免因缺氧或血液中二氧化碳過高或低及腦部灌流不足所致之腦部二次傷害。當腦部情況相對穩定後，便進入脫離呼吸器的訓練階段。脫離呼吸器的標準可由各醫院自行訂定，一般而言，是漸近式減少呼吸器的輔助，並經測試通過可脫離呼吸器的標準後才拔除呼吸管。由於神經重症病患的首要目標是神經功能的恢復，因此太早勉強脫離呼吸器可能增加心肺負擔、增加重插管機會、延後腦部復原，反之太晚脫離呼吸器則會增加使用呼吸器的併發症如: 肺炎、延後接受復健的時程、增加醫療資源耗用。神經重症病患有高於其它重症病患的再插管率，也有較高的比率使用呼吸器14天以上需要接受氣切治療。原因可能為腦部手術後常合併意識不清及呼吸肌肉功能不足，使這類病患對呼吸器依賴性高，因此，並不適合與其它重症病患使用相同的評估標準來決定拔管時機。 本研究將分析北部某醫學中心外科及創傷加護病房中五年內至少400位接受腦部手術後使用呼吸器超過48小時之病患資料，使用邏輯斯迴歸及資料探勘分析找出影響脫離呼吸器後72小時內重插管及使用呼吸器超過14天之相關因子並建構其預測模型，探討臨床上不同時間點之病患意識、過去病史、生命徵象、檢驗值、呼吸器脫離評估指標與呼吸器脫離成敗之間的關連性。<br> Abstract: Neurocritical care is the intensive care management of patients with life-threatening neurological and neurosurgical illnesses such as ischemic stroke, intracerebral hemorrhage, subarachnoid hemorrhage, and brain trauma. Following surgery for these diseases, patients require ventilator support. Cautious management of mechanical ventilation is required to avoid secondary injury from hypoxemia, hypo-or hypercarbia, or decreases in cerebral perfusion pressure. When the intracranial conditions get resolved, weaning process starts. Delayed weaning can lead to complications such as ventilator induced lung injury, ventilator associated pneumonia, and delayed rehabilitation. On the other hand, premature weaning can lead to complications like loss of the airway, aspiration and respiratory muscle fatigue, resulting in reintubation. Weaning protocol and parameters differs in different hospitals. Neurocritcal patients usually experience higher reintubation rate and prolonged mechanical ventilation.In this study, we plan to analyze factors associated with reintubation and prolonged mechanical ventilation using logistic regression and data mining analysis. Data, including consciousness, past history, lab data, and weaning parameters, of neurocritical patients who used mechanical ventilation for more than 48 hours will be collected. Predictive models will be built for reintubation and prolonged mechanical ventilation.預後腦中風頭部外傷邏輯斯迴歸資料探勘data mininglogistic regressionoutcomestroketraumatic brain injury