加速失敗時間模式分析新發乳癌病患併發血栓栓塞對其存活的影響

Accelerated failure time model analysis of the impact of thromboembolic events on survival in newly diagnosed breast cancer patients

目標：血栓栓塞併發症（TEEs）為乳癌併發症主要之一且增加死亡風險，本研究擬探討台灣新發乳癌病患併發血栓栓塞之相關因素及對其兩年存活的影響。方法：本回溯性世代研究使用1997至2010年全民健康保險研究資料庫。以Poisson分佈探討2000到2008年新發乳癌世代觀察兩年其併發TEEs的發生情形。透過propensity socre匹配後再以Accelerated Failure Time Model檢視有無併發症病患對其兩年存活之影響。結果：2000－2008年新發乳癌世代共80,598人，罹癌確診後兩年內併發TEEs之累積發生率1.78％（95％ CI=1.74－1.82），且第一個半年併發症發生密度為1.38發生事件數／每100個人年。併發TEEs危險因子有年齡（≧65歲比＜45歲adjusted incidence rate ratio[adj.IRR]=3.87，95％ CI=3.80－3.93），共病指數（≧3分比0分adj.IRR=1.96，95％ CI=1.93－2.00），有高血壓病史（adj.IRR=1.43，95％ CI=1.41－1.45），有乳房手術者（adj.IRR=1.47，95％ CI=1.45－1.49），接受放射線治療者（adj.IRR=1.29，95％ CI=1.28－1.31），接受化學藥物治療者（adj.IRR=1.40，95％ CI=1.38－1.42），醫院層級（地區醫院比醫學中心adj.IRR=1.13，95％ CI=1.11－1.15）。AFT model分析顯示在控制其他變項後，有併發TEEs者比無者有較高比例的兩年死亡風險（adjusted Rate Ratio =2.55，95％ CI=2.24－2.91），且風險隨時間增加而增加。結論：新發乳癌病患併發TEEs會增加其死亡風險。醫療專業人員者可針對易併發TEE之危險群（如年長、共病指數高、有高血壓、接受乳癌手術、接受化療放療），即早警覺並提供適當的預防措施以利降低併發症的發生及提升癌症照護品質。

Objectives: Thromboembolic events (TEEs) are major complications affecting breast cancer patients, and are associated with increased mortality. The purpose of this study was to determine the incidence and risk factors for developing a new TEE in a population-based cohort of breast cancer patients, and also to determine the effect of TEEs on 2-year survival. Methods: A retrospective cohort study was conducted between 1997- 2010 using the National Health Insurance Research Database in Taiwan. The incidence of TEEs amongst 2000-2008 breast cancer cohorts and the associated risk factors were determined based on the Poisson distribution. Propensity score matching was used to determine the effect of TEEs on 2-year survival through accelerated failure-time (AFT) model analysis. Results: Amongst 80,598 patients with breast cancer between 2000 and 2008, the 2-year cumulative TEE incidence was 1.78%, with a rate of 1.38 events/100 patient-years during the first 6 months. Based on a multivariate model, significant predictors of developing TEEs within 2 years were age (adjusted incidence rate ratio [aIRR] = 3.87 if ≧ 65 years vs. <45 years; 95% CI = 3.80-3.93), co-morbidity (aIRR = 1.96 if ≧3 vs. 0, 95% CI = 1.93- 2.00), hypertension (aIRR = 1.43 if yes vs. no, 95% CI = 1.41-1.45), history of breast surgery (aIRR = 1.47 if yes vs. no, 95% CI = 1.45-1.49), history of radiotherapy (aIRR = 1.29 if yes vs. no, 95% CI = 1.28-1.31), history of chemotherapy (aIRR = 1.40 if yes vs. no, 95% CI = 1.38- 1.42), and type of hospital (aIRR = 1.13 if district medical center vs. patient medical center 95% CI = 1.11-1.15). Based on AFT model analysis, and after controlling for other variables, patients with TEEs had a higher risk of 2-year death than those without TEEs (aRR = 2.55, 95% CI = 2.24-2.91) and had an increased hazard ratio over time. Conclusions: Breast cancer patients with TEEs have a higher risk of death. Early detection of risk factors of TEEs, such as older age, more co-morbidities, hypertension, and histories of surgery, chemotherapy, and radiotherapy) and appropriate preventive care should be provided by health professionals for decreasing the complications and improving the quality of cancer care.