安寧病房老年癌末病人14日內存活期之預測－－依據病人的臨床特徵及實驗室檢查

Prediction of Mortality within 14 days for Elderly Terminally Ill Cancer Patients by Clinical Characteristics and Laboratory Tests

背景及目的︰老年癌末病人的存活期是病人、家屬及醫療人員所關心的議題。若能較準確地預測病人之存活期，病人將會有較適宜的末期生涯規劃及治療上的考量，家屬對未來事物也會有較適宜的規劃，醫療人員則能明智地使用醫療及避免病人在臨終時的不必要治療。本研究目的在分析住進安寧病房之老年癌末病人住院後至死亡小於或等於14日的預測因子。方法︰本研究為病歷回顧法，收集2004年11月至2007年5月間，癌末住院之老年病人作為研究對象，登錄病人第一次入住安寧病房之病歷內所記載病人住院後24小時內的基本人口學資料、臨床特徵、實驗室檢查結果及病人的死亡日期等資料。結果︰共433位老年癌末病人納入分析，住院後至死亡時間小於或等於14日者有187位(43.2%)，大於14日者有246位(56.8%)。病人的平均年齡為74.7±6.3歲，存活中位日數為17日。兩組間之比較，在存活小於或等於14日組，有較多的病人是男性、較差的身體功能、較多病人罹患肝癌、較高的實驗室檢查結果包括WBC、BUN、creatinine、SGOT、total bilirubin、K、corrected Ca、及較低的實驗室檢查結果包括lymphocyte比例、albumin。多因子分析結果，顯著的預測因子包括︰男性(p=0.002)、身體功能差者(p<0.001)、癌症轉移器官之總數較多(p=0.020)、癌症原發部位為肝臟者(p=0.004)、WBC較高(p=0.010)、及BUN較高(p<0.001)。進一步計算老年病人住院後至死亡時間小於或等於14日之機率，其準確性為76.5%。結論︰藉本研究發現的男性、身體功能狀態、癌症轉移之器官總數、癌症原發部位為肝臟者、WBC、及BUN等預測因子，可提供醫護人員對老年癌末病人存活期的預測參考。

Background: The issue of survival time for elderly, terminally ill cancer patients is always a major concern for patients, families, and the hospice-care team. If we could provide accurate prognostic information, then patients could benefit from appropriate planning for end-of-life, and the patient's family from appropriate planning about their daily work. Physicians can optimize our care by avoiding futile treatment measures during the patient's dying stage. Objectives: Our aim was to analyze the different factors in predicting mortality within 2 weeks of hospice admission for elderly cancer patients. Methods: This retrospective study was conducted on elderly patients who were admitted to the hospital-based hospice ward during the period of November 2004 to May 2007. Data were collected by chart review and included demographic data, clinical characteristics, laboratory tests within 24 hours after admission, and the date of death. Results: Of 433 patients who were enrolled in the study, 187 died within 14 days after hospice admission and 246 patients died after more than 14 days. When we compared the 2 groups, we found that some factors were significant in the group who died within 14 days after admission. These included male gender; worse performance status (ECOG=3 or 4 vs. =1 or 2); primary cancer in the liver; higher values for WBC, BUN, creatinine, SGOT, total bilirubin, potassium, corrected calcium, and lower values for lymphocyte percentage and albumin. On logistic regression analysis, the significant factors for predicting death within 14 days after hospice admission in elderly, terminally ill cancer patients were male gender (p = 0.002), worse performance status (p < 0.001), a greater number of metastases (p = 0.020), liver as the primary cancer site (p = 0.004), more WBCs (p = 0.010) and higher BUN (p < 0.001). The accuracy of this model was 76.5% in this study. Conclusion: We therefore concluded that the death of elderly, terminally ill cancer patients could be predicted based on our protocol. Our study revealed significant prognostic factors to be male gender, performance status, number of metastatic sites involved, hepatocellular carcinoma, WBC, and BUN. As a result, we can deliver more appropriate and compassionate care to these patients and their families.