北京大学肿瘤医院残胃上的癌临床病理特征及预后相关因素分析

Abstract

目的 探讨首次疾病为良性患者与首次疾病为恶性患者残胃发生癌变的间隔时间、临床病理特点及预后相关因素.方法 以日本胃癌协会2017年提出的残胃上的癌(CRS)定义为标准,回顾性分析自1992年3月至2017年3月就诊于北京大学肿瘤医院胃肠肿瘤中心的CRS患者临床资料.首次疾病为良性(CRS-B)与首次疾病为恶性(CRS-M)两组患者组间的计量资料采用独立样本t检验或Mann-Whitney U非参数检验;组间计数资料采用χ2检验或Fisher精确概率法.采用Spearman相关分析检验变量间的相关性.采用Kaplan-Meier进行生存分析,Cox回归模型分析影响患者预后的风险因素,采用受试者工作特征曲线(ROC曲线)下面积来比较不同肿瘤分期标准区分患者预后的准确性.结果 共计有89例患者纳入研究,男女比为5.4:1.0.其中CRS-B组46例,男女比例为14.3:1.0;CRS-M组43例,男女比例为2.9:1.0;两组差异有统计学意义(χ2= 6.091,P =0.019).CRS-B和CRS-M两组患者术后残胃发生癌变的间隔时间分别为342(36~576)月和47(12~360)月,差异有统计学意义(t =8.887,P=0.000).CRS-B组患者残胃发生癌变的间隔时间与初次手术的手术方式存在一定相关性(相关系数为0.398,P=0.006);CRS-M组则与初次手术时患者年龄存在一定相关性(相关系数为0.337,P = 0.027).当定义胃癌术后新生癌先后两次手术病理结果不同时:复发癌27例,新生癌15例;两组患者残胃癌变的间隔时间分别为46(12~132)月和60(12~360)月,差异有统计学意义(t= 5.652,P=0.023).CRS-B组患者CRS发生于胃肠吻合口区、胃壁吻合或闭合区以及非吻合口区分别占60.9%(28/46),23.9%(11/46)和15.2%(7/46);而CRS-M组患者其CRS发生于胃肠吻合口区、胃壁吻合或闭合区以及非吻合口区则分别占39.5%(17/43)、16.3%(7/43)和44.2%(19/43),差异无统计学意义(χ2=4.726,P=0.096).在77例接受根治性手术的患者中,总体并发症发生率为20.8%(16/77),主要为感染性并发症(8例)及呼吸系统并发症(7例).CRS-M组和CRS-B组患者3年生存率分别为78.4%和62.6%,差异有统计学意义(χ2=3.969, P=0.046).淋巴结阳性比率(LNR)与肿瘤N分期对不同肿瘤分期标准区分患者预后的ROC曲线下面积分别为0.725和0.639.多因素分析显示,病理T分期(HR=1.192,95%CI:1.032~1.376, P=0.017)为影响CRS患者预后的独立危险因素.结论 CRS存在男性多于女性的性别差异;CRS-B患者残胃癌变的间隔时间明显长于CRS-M者,CRS-B残胃发生癌变的间隔时间与初次手术方式相关,CRS-M则与初次手术时的患者年龄相关;CRS-B残胃主要癌变区域集中于胃肠吻合口区,CRS-M则主要集中于非吻合口区域.病理T分期是CRS患者的独立预后因素.

Objective To investigate the interval time to canceration, clinicopathological characteristics and prognostic factors of carcinoma in remnant stomach(CRS)in patients with primary benign diseases or primary malignant tumors. Methods Based on the criteria of the definition of CRS proposed by Japanese Gastric Cancer Association in 2017, a retrospective analysis was conducted on clinicopathological characteristics of patients diagnosed with CRS at Peking University Cancer Hospital from March 1992 to March 2017. Between patients with primary benign diseases (CBS-B group) and primary malignant tumors(CBS-M group), continuous variables were compared using the Student's t-test or the Mann-Whitney U test; categorical variables were compared using the chi-square test or Fisher's exact test. Spearmen-Rho was used to examine correlation. Survival was estimated and compared using Kaplan-Meier methods. Cox proportional hazards regression was applied to identify independent prognostic factors. Area under ROC curve(AUC)was used to evaluate and compare prediction accuracy. Results A total of 89 patients were included in the study with a male: female ratio of 5.4:1.0. The male: female ratio in CRS-B(n=46)and CRS-M(n=43)group was 14.3:1.0 and 2.9:1.0 respectively with significant difference(χ2= 6.091, P = 0.019). The interval time to canceration in CRS-B and CRS-M group was 342(36-576)months and 47(12-360)months respectively with significant difference(t = 8.887, P = 0.000). The interval time to canceration was correlated with the first operative procedure in CRS-B group(r=0.398, P=0.006), while interval time to canceration was correlated with the age at the first operation in CRS-M group(r=0.337,P=0.027). After differentiating the pathological findings of the first operative sample and the second operative sample, 27 patients presented recurrence and 15 patients had new cancer, and the corresponding interval time to canceration was 46(12-132)months and 60(12-360)months respectively with significant difference(t = 5.652, P=0.023). In CRS-B group, location of stump carcinoma in gastric intestinal anastomosis, gastric anastomosis, and non-anastomosis area was found in 60.9%(28/46), 23.9%(11/46)and 15.2%(7/46)respectively, and the corresponding percentage in CRS-M group was 39.5% (17/43), 16.3%(7/43)and 44.2%(19/43)respectively without significant difference(χ2= 4.726, P=0.096). Among 77 patients with radical gastrectomy, the overall surgical complication rate was 20.8%(16/77), including 8 cases of infection and 7 cases of respiratory system diseases. The 3-year survival rate was 78.4% and 62.6% in CRS-B and CRS-M group respectively with significant difference(χ2= 3.969, P=0.046), indicating better prognosis of CRS-B patients. The AUC for the lymph nodes ratio and N staging was 0.725 and 0.639 respectively. Multivariate analysis showed the pathological T staging was an independent risk factor of prognosis(HR = 1.192, 95%CI:1.032-1.376, P = 0.017). Conclusions Men have more CRS than women. The interval time to canceration is correlated to the first operative procedure for CRS-B patients, while it is correlated to the age at the first operation for CRS-M patients. The major location of CRS is in the gastrointestinal anastomosis for CRS-B patients and in non-anastomosis area for CRS-M patients. Main postoperative complications include respiratory and infectious complications. Pathological T staging is an independent prognostic risk factor for CRS patients.