明智就醫: 口腔癌患者治療後追蹤計劃與預後之相關性-以中部某醫學中心為例

Abstract

研究背景與目的：根據臺灣癌登統計，以男性而言，口腔癌之發生率及死亡率均排名第三，為現階段臺灣最常見之頭頸癌。目前對於口腔癌病人治療後之追蹤計劃沒有一定的標準且仍欠缺共識，雖然大部份針對頭頸癌治療後追蹤計劃之研究指出影像學檢查及門診追蹤頻次與頭頸癌病人預後沒有統計學上之相關，唯大部份均不是專門針對口腔癌之研究。近來明智就醫之議題愈來愈受到重視，旨在促進病人與臨床醫師間的對話，避免不必要的醫學檢查、治療和程序。所以本研究之目的在於探討口腔癌病人，在接受完整治療後，不同影像學檢查頻率及門診次數對於預後之影響。 研究方法：本研究採回溯性之病歷回顧，利用臺中榮民總醫院之臨床資料庫之口腔癌病人，探討追蹤期間、電腦斷層/磁振造數次數、腹部超音波次數、骨骼掃瞄次數等變項對於預後之影響。分析方法採描述性統計及推論性統計呈現，存活分析以Kaplan-Meier法為之，不同組別之比較以對數－等級檢定 (log-rank test) 分析，影響存活的因素 (單變項及多變項分析) 以Cox比例風險模式(Cox proportional hazards model) 為之。 研究結果：總共有741位人納入最終分析，平均年齡為54.8歲，男性佔大多數 (655位，88.4%)，最常見之原發部位為舌癌 (338位，45.6%)，早期及晚期病人分別為391位(52.8%) 及350位 (47.2%)，在追蹤期間有233位 (31.4%) 出現局部或區域性復發，166位 (22.4%) 在追蹤期間因口腔癌因素死亡，平均追蹤期間為47.9個月。追蹤期間每人平均接受電腦斷層/磁振造影為 5.0次，肝臟超音波為 2.9次，骨骼掃瞄為 2.3次，正子攝影為 1.6次，門診次數平均為42.3次。在多變項分析當中，整體而言，電腦斷層/磁振造影頻次、肝臟超音波頻次、及骨骼掃瞄頻次與口腔癌病人之疾病別死亡沒有統計學上之相關，但正子攝影與門診頻次與口腔癌病人之預後呈負相關。如將口腔癌病人根據期別及年齡分為兩組，我們發現在第三四期的口腔病人，較多的電腦斷層/磁振造影檢查 (>=1.3 次/年) 有較低之死亡風險 (死亡風險：0.52，95%信賴區間：0.34-0.81)，而在大於等於54歲的口腔癌病人族群，較多之電腦斷層/磁振造影 (>=1.3 次/年) (死亡風險：0.51，95%信賴區間：0.29-0.90) 及肝臟超音波 (>=0.66 次/年) (死亡風險：0.56，95%信賴區間：0.32-0.99)，都能減少口腔癌病人之死亡風險。 結論：口腔癌病人治療後的追蹤計劃應朝精準化之方向前進，針對早期及年輕 (小於54歲) 的口腔癌病人治療後，可依病人之症狀及醫師臨床檢查的結果來安排影像學檢查，但在晚期或年長 (大於等於54歲) 的口腔癌病人治療後，建議至少每年定期安排影像學之檢查，以期能早期發現復發，及早介入來提昇存活率。未來可運用多機構之資料 (如：健保資料庫) 來進一步探入分析。另也可以設計前瞻性之隨機分派研究，與其他機構進行多中心之臨床實驗來釐清影像學檢查與口腔癌病人預後的相關性。

Background & Purpose: According to the data from the Taiwan Cancer Registry, the incidence and mortality rate of oral cavity cancer patients all ranks 3rd among all cancers in male population. In addition, oral cavity cancer is the most prevalent head and neck cancer nowadays in Taiwan. There is no standard or consensus about the follow-up protocol after comprehensive treatment of oral cavity cancer patients. Although most studies about follow-up protocol after treatment of head and neck cancer patients indicated that there is no association between frequency of image studies and prognosis, few studies focused solely on oral cavity cancer patients. Choosing wisely is a current trend and its goal is to promote patient-physician conversations about unnecessary medical tests and procedures. The aim of this study was to investigate the relationship between the prognosis of oral cavity cancer patients and the frequency of image studies as well as the frequency of clinic visits. Methods: This is a retrospective study which was approved by the Institutional Review Board of Taichung Veterans Hospital (TCVGH). We used clinical database of TCVGH to explore oral cavity patients who were treated in TCVGH. We collected demographic data, follow-up duration, frequency of Computerized Tomography (CT)/Magnetic Resonance Imaging (MRI), abdominal sonogram, whole body bone scan, and clinic visits of oral cavity cancer patients after comprehensive treatment. Then we analyzed the association between abovementioned variables and prognosis of oral cavity cancer patients. Demographic data were presented by descriptive statistics and comparison between subgroups was analyzed with inferential statistics. Survival analysis was done by Kaplan-Meier method and comparison between different subgroups was analyzed by log-rank test. The variables related to prognosis of oral cavity cancer patients were examined by Cox proportional hazards model. Results: There were totally 741 oral cavity cancer patients enrolled into final analyses. The average age was 54.8 years and male accounted for most of the patients (n=655, 88.4%). In addition, the most common primary site was tongue (n=338, 45.6%). Patients with early stage disease and late stage disease were 391 (52.8%) and 350 (47.2%), respectively. Two hundred and thirty-three patients (31.4%) developed loco-regional recurrence while 166 patients (22.4%) died because of oral cavity cancer during follow-up period. Besides, the average follow-up period was 47.9 months. During follow-up period, the average numbers of examination per person for CT/MRI, liver sonogram, whole body bone scan, and Positron Emission Tomography (PET) were 5.0, 2.9, 2.3, and 1.6, respectively. The average clinic visits was 42.3 visits. In multivariate analysis, there were no significant statistic difference between prognosis of oral cavity cancer patients and image studies except PET scan. There was a negative correlation between prognosis of oral cavity cancer patients and frequency of PET scan after treatment. Also, more clinic visits was correlated with a poor prognosis of oral cavity cancer patients. We then divided all patients into subgroups based on tumor stage and age. We found more CT/MRI (>=1.3 times/year) was associated with a better prognosis in late stage oral cavity cancer patients (Odds ratio [OR] for mortality: 0.52，95% confidence interval [CI]: 0.34-0.81). In elder patients, disease-specific mortality rates were lower in patients receiving more CT/MRI (>=1.3 times/year) (OR: 0.51, 95% CI: 0.29-0.90) as well as more liver sonogram (>=0.66 times/year) (OR: 0.56, 95% CI: 0.32-0.99). Conclusion: Post-treatment surveillance of oral cavity cancer patients should be more precisely. Based on our study, we do not recommend routine image study in asymptomatic early and younger oral cavity cancer patients after comprehensive treatment. However, annually image study is recommended in elder or late stage oral cavity cancer patients in order to detect loco-regional recurrence in advance. Further large-scale study is warrant to clarify the relationship between prognosis of oral cavity cancer patients and post-treatment surveillance image studies.