糖尿病對肝硬化合併肝癌病患接受肝切除手術預後之影響

Impact of Diabetes on the Surgical Outcome Following Hepatic Resection in Cirrhotic Patients with Hepatocellular Carcinoma

丁金聰(Chin Tsung, Ting)

指導教授 ： 郭乃文

共同指導教授 ： 陳潤秋

臺北醫學大學/公共衛生暨營養學院/醫務管理學研究所/碩士(2010年)

https://doi.org/10.6831/TMU.2010.00193

摘要

研究背景：
肝癌是現今世界癌症死亡原因的第五位。在台灣，肝癌也是僅次於肺癌而位居十大癌症死亡原因的第二位。行政院衛生署國民健康局為提昇癌症的預防與治療品質，因此著手推行｢癌症治療品質提昇計劃｣，希望能提昇所有癌症的防治。 2004 年初步推行的｢癌症治療品質提昇計劃｣共規劃與執行了包括肝癌等的六大癌症的品質提昇計劃，肝癌尤其是此六大癌症之首，而且也是六大癌症中最為複雜與困難的項目。｢癌症治療品質提昇計劃｣的兩大重點為癌症防治指標與指引的建立，為建立一個適合國人的肝癌防治指標與指引，國民健康局於 2010 年委由臺灣肝癌醫學會與各大醫學中心共同完成此一任務。因此，基於此任務下進行的｢發展癌症首次治療型態指標｣研究計劃，目標設定為訂定全國適用的肝癌首次治療型態指標，並成立了｢指標發展委員會｣進行肝癌指標的研究。
影響肝癌診斷、治療與追蹤的指標包括肝功能指標、肝癌的腫瘤指標與新陳代謝疾病指標三大類。尤其是包括糖尿病、肥胖症等疾病的新陳代謝疾病指標更是重要，因為糖尿病在肝炎、肝硬化與肝癌病患的流行率明顯的比較高，對於肝臟的代謝功能會造成干擾，而且許多研究也發現糖尿病對肝癌的發生、治療後的肝癌復發與治療的長期預後都有影響，因此糖尿病可能可以視為影響肝癌治療的重要指標。本研究的目的就是希望能以接受手術治療後的肝癌病患為樣本，分析糖尿病指標對肝癌病患接受手術治療後，在肝癌復發、長期預後與醫療耗用的影響，以了解糖尿病指標是否應該列入肝癌防治的指標，並提供治療指引建立的參考。

研究方法：
本研究為一回溯性的準實驗研究，收集了台北市某區域教學醫院接受手術的肝癌病患為研究對象，從 2001 年 1 月起，至 2008 年 12 月為止，於此醫院接受肝臟切除手術的肝癌病患共有 459 位，其中有 389 位肝癌病患因接受完整的追蹤治療因而進入本研究分析之內，包括男性有 281 位，女性 108 位，平均年齡為 61.7 歲。研究架構以手術後的復發時間、手術後的總存活時間與總住院日數為依變項，以包括性別、年齡、肝炎的類型、血小板貧血症、血清胎兒蛋白指數、血清 GOT 指數、BMI 值、Child 的肝功能量表分期、腫瘤大小、腫瘤的數目、肝癌手術切除的安全距離、肝癌的細胞分化、是否合併血管侵犯等為控制變項，分析 (1) 是否罹患糖尿病與 (2) 糖尿病的控制方法兩個自變項在依變項的統計差異。因此研究流程採兩階段設計，第一階段將所有接受手術治療的肝癌病患分為 Group A (沒有罹患糖尿病的肝癌病患族群)與 Group B (罹患糖尿病的肝癌病患族群)兩組，比較這兩組病患在基本特性與依變項的差異。第二階段將所有罹患糖尿病且接受手術治療的肝癌病患分為 Group a (使用口服降血糖

藥物且罹患糖尿病肝癌病患族群) 、Group b (使用胰島素針劑且罹患糖尿病肝癌病患族群) 二組，同樣比較這兩組病患在基本特性與依變項的差異。基於以上的研究設計，本研究的假說為：
(1) 糖尿病指標對肝癌病患接受手術治療後的復發時間沒有顯著影響。
(2) 糖尿病指標對肝癌病患接受手術治療後的總存活時間沒有顯著影響。
(3) 糖尿病指標對肝癌病患接受手術治療後的總住院日沒有顯著影響。
(4) 糖尿病的控制方法對於罹患糖尿病與肝硬化的肝癌病患接受肝臟切除手術後的復發時間沒有顯著影響。
(5) 糖尿病的控制方法對於罹患糖尿病與肝硬化的肝癌病患接受肝臟切除手術後的總存活時間沒有顯著影響。
(6) 糖尿病的控制方法對於罹患糖尿病與肝硬化的肝癌病患接受肝臟切除手術前後的總住院日數沒有顯著影響。
本研究分析的最終目的在於檢驗以上六個假說是否成立，統計分析的工具使用的是 SPSS for Windows, 16.0 版，在病患基本特性與總住院日數的統計方法採取 independent two sample t-test 或 χ2 test，復發時間與總存活時間的統計方法則使用Kaplan-Meier analysis (log-rank test) 與 Cox regression analysis。包括單變異數分析與多變異數分析，統計學的統計明顯差異定義為 P<0.05。

研究結果：
第一階段的病患基本特性分析發現：罹患糖尿病的肝癌病患明顯的手術前的血糖控制較差、肥胖症的比例較高、多發性肝癌的比例略高與當次手術的總住院日較長。而且存活分析也發現是否罹患糖尿病與 Child 的肝功能量表分期、血清 GOT 指數共同為影響肝癌病患接受手術治療的重要指標。顯然本研究對於糖尿病指標與依變項的三個假說皆不成立，糖尿病指標對於肝癌病患接受手術治療後的復發時間、總存活時間與總住院日數皆有顯著影響。
第二階段針對罹患糖尿病且接受手術治療的肝癌病患作分析，分析發現不管是使用口服降血糖藥物控制血糖的肝癌病患，或是使用胰島素針劑控制血糖的肝癌病患，病患的基本特性比較幾無差異。而存活分析顯示，比較使用口服降血糖藥物與使用胰島素針劑兩組病患的手術後肝癌復發比率與總住院日並沒有明顯的不同。而使用口服降血糖藥物以控制血糖的病患比起使用胰島素降血糖針劑以控制血糖的病患，手術後的存活時間比較長。因此，本研究的假說：糖尿病的控制方法對於罹患糖尿病的肝癌病患接受肝臟切除手術後的復發時間與總住院日沒有顯著影響是成立的，而糖尿病的控制方法對於合併糖尿病與肝硬化的肝癌病患接受肝臟切除手術後的總存活時間沒有顯著影響是不成立的。

結論：
糖尿病指標因為對於肝癌病患接受肝臟切除手術後的復發時間與存活時
間皆有影響，因此應該在｢發展癌症首次治療型態指標｣研究計劃中列為肝癌治療的指標，以為臺灣肝癌醫學會與國民健康局在建立肝癌治療指引的參考，對於肝癌病患接受手術治療的安全與品質提昇必有相當大的助益。
雖然糖尿病的控制方法對於罹患糖尿病的肝癌病患接受肝癌切除手術後的復發時間與總住院日的影響並不顯著，僅只對肝癌切除手術後的總存活時間略有影響。但我們也注意到罹患糖尿病的肝癌病患與沒有罹患糖尿病的肝癌病患的血糖控制明顯的有差異存在，顯然醫師必須更深入研究糖尿病與肝臟代謝功能的交互影響，並對罹患糖尿病的肝癌病患採取更積極的血糖控制策略，方可使罹患糖尿病的肝癌病患治療效果更好。
糖尿病在醫療上的資源耗用也是一個嚴重的問題，就以 1998 年的全民健康保險總支出為例，糖尿病的醫療耗用比例就佔了全民健康保險總支出的 11.5%，而且這幾年來糖尿病患的流行率也逐年上升，相對醫療的耗用比率也就隨之上升。就以本研究所發現，罹患糖尿病的肝癌病患接受手術治療後的總住院日比較於沒有罹患糖尿病的肝癌病患顯著的比較長。而在總住院日的比較，糖尿病的控制方法對於肝癌病患接受肝臟切除手術前後的總住院日數則沒有影響。因此罹患糖尿病的肝癌病患接受治療的醫療花費無可避免的比較高。因此，強化糖尿病病患的追蹤與血糖控制的意義將不僅僅只是醫療上提高存活率與減少併發症而已，在醫療耗用的管控也別具重要的意義。

關鍵字

肝癌 ； 癌症品質提昇計劃 ； 糖尿病

並列摘要

Background:
Hepatocellular carcinoma (HCC) is the fifth most common malignancy in the world. Its prevalence is particularly high in Asia and Africa. In Taiwan, HCC is considered the second common malignancy, which frequently causes deaths. The etiology of HCC is not clearly understood. Viral hepatitis, such as hepatitis B infection and hepatitis C infection, is the most common cause of HCC. Some newly found factors, such as alcoholic liver disease, non-alcoholic steatohepatitis (NASH) and Aflatoxin, also serve as less frequent etiologies of HCC. Recent studies suggest that the coexistence of diabetes is a risk factor for developing HCC as well. However, it is controversial regarding whether diabetes is an independent risk factor for post-treatment recurrence and the outcome of HCC. The aim of this study was to analyze the impact of diabetes on the postoperative recurrence as well as the surgical outcome after hepatic resection in patients with liver cirrhosis and HCC.

Method:
From January 2000 to December 2008, a total of 461 cirrhotic patients who had undergone curative resection for HCC were treated at a hospital located in Taipei. Among these 461 patients, there were 389 cirrhotic patients have received regular post-operative follow-ups. For these patients, there were 227 (70%) patients classified as non-diabetes group (group A), and 117 (30%) patients as diabetes group, of which has abnormal serum blood sugar and diabetes history (Group B). The patient characteristics, recurrent-free survival rates, overall survival rates and total hospital stays after surgical treatments between these two groups were examined. The factors contributing to recurrence and long term surgical outcomes of these two groups were also analyzed.
The diabetes group (Group B) was classified into “diabetes group with oral hypoglycemic agent control group” (group a) and “diabetes group with insulin control group” (group b). I analyzed the patient characteristics, recurrent-free survival rates, overall survival rates and hospital stays after surgical treatments between these two groups as well.

Result:
In terms of patient characteristics between “Group A” and “Group B”, Group B has significant higher serum blood sugar levels, more obese and more frequently prevalent of non-viral hepatitis than those of Group A. Results of Kaplan-Meier survival analysis of post treatment surgical outcomes between Group A and Group B suggested that the recurrence-free survival rates of Group B were significantly higher than those of Group A (P=.001). The overall survival rates of Group B were significantly higher than those of Group A (P = 0.01). However, patients with diabetes (Group B) showed significantly higher post-operative recurrent rates, poor long term surgical outcomes and longer hospital stays comparing to those of the non-diabetes group (Group A).
Furthermore, the post-operative recurrent rates, overall survival rates and total hospital stays between the two subgroups of diabetic patients were also analyzed. I compared the surgical outcomes between these two groups of patients with Kaplan-Meier survival analysis as well. Results showed that the diabetic patients with insulin control group (group b) had poor long term prognosis. Other than these, there were no significant differences for recurrent-free survivals and total hospital stays between group a and group b.

Conclusion:
Results of this study suggested that diabetes is a critical risk factor for the recurrence of cirrhotic patients combined with HCC underwent surgical treatments. Patients with diabetes also have decreasing long term overall survivals in terms of surgical outcomes following hepatic resection in cirrhotic patients with HCC. Therefore, these results suggest that diabetes should be treated as an indicator for HCC therapeutic guideline.
In addition, this study found group b has poor surgical outcomes compared with group a. There were no other significant differences between recurrence-free survival rates and hospital stays between group a and group b. Results also show that diabetic HCC patients had higher serum glucose level. However, there were no significant differences in terms of serum glucose level between group a and group b. These results indicated that the diabetic HCC patients had poor sugar control compared with the general population. Insulin therapy does not have better efficacy in comparison to oral hypoglycemic agent for blood sugar control. A more aggressive blood sugar control for diabetic cirrhotic HCC patients is needed. Better blood sugar controls may result a better outcome for diabetic cirrhotic HCC patients.
Furthermore, results show that diabetic cirrhotic HCC patients (Group B) have longer post-operative hospital stays than non-diabetic cirrhotic HCC patients (Group A). This implies that Group B may have higher healthcare costs than that of Group A. There is no significant difference between group a and group b in terms of post- operative hospital stays. These results implied that the serum blood sugar control may dismiss related cost, but is not correlated to method of blood sugar control. Adequate blood sugar control can decrease the total cost.
In conclusion, this research suggests that diabetes could be served as an independent risk factor for the recurrences and long term surgical outcomes in cirrhotic HCC patients who received hepatic resections. Diabetic cirrhotic HCC patients also may incur higher surgical costs than non-diabetic cirrhotic HCC patients. The cause of higher surgical costs may be resulted from inadequate blood sugar control. More aggressive blood sugar control can achieve a lower post-operative recurrence rates, better long term surgical outcomes and lower surgical costs. Cirrhotic HCC patients who also suffer from diabetes should be closely followed for postoperative recurrences and prognosis.

並列關鍵字

Hepatocellular carcinoma ； Diabetes mellites

參考文獻

行政院衛生署國民健康局網站

recurrence in human small hepatocellular carcinoma. Gastroenterology. 108,

American Diabetes Association. (2010) Diagnosis and classification of diabetes mellitus. Diabetes care, 33(supp.1), 62-69.

Anwar W.A., Khaled H.M., Amra H.A., Loffredo C.A. (2008) Changing pattern of hepatocellular carcinoma (HCC) and its risk factors in Egypt. Mutat Res. 659, 176-84.

Bart J. Veldt,1 Wendong Chen, et al. (2008) Increased Risk of Hepatocellular Carcinoma Among Patients with Hepatitis C Cirrhosis and Diabetes Mellitus. Hepatology. 47,1856-1862.