L3骨骼肌质量指数与控制营养状况评分对胃癌患者术后预后的预测价值.pdf

现代肿瘤医学2 0 2 3年11月第31卷第2 1期MODERN ONCOLOGY,Nov.2023,VOL.31,No.213985.L3骨骼肌质量指数与控制营养状况评分对胃癌患者术后预后的预测价值钱雪蔚，朱琴琴，江晓晖南通大学附属肿瘤医院胃肠外科，江苏南通2 2 6 0 0 0【摘要】目的：探讨第三腰椎骨骼肌质量指数(L3-skeletal muscle mass index,L3-SMI)和控制营养状况（c o n t r o l l i n g n u t r i t i o n a l s t a t u s，CO NU T）评分对胃癌患者术后预后的预测价值。方法：收集2 0 17 年0 1月至2020年12 月于我院行胃癌手术的52 8 例患者的临床及生存资料，采用受试者工作特征（receiver operatingcharacteristic,ROC)曲线确定不同性别患者术前 L3-SMI的最佳截断值，比较不同 L3-SMI、CO NU T 评分水平患者的临床病理特征。使用 Kaplan-Meier 法进行生存分析,采用 Cox比例风险模型进行单因素和多因素回归分析胃癌患者术后预后的影响因素。应用R软件构建胃癌患者术后总生存时间（overall survival,OS)列线图预测模型,计算C指数评估模型的准确性,绘制校准图评价模型的预测效能。结果：术前 L3-SMI预测胃癌患者预后的最佳截断值：男性为43.95cm/m、女性为37.6 cm/m，曲线下面积（area underthe curve,AUC)值分别为0.7 2 9(9 5%CI为0.6 7 2 0.7 8 5)和0.6 8 0(9 5%CI为0.6 0 7 0.7 53）。术前低L3-SMI水平胃癌患者1、3、5年累积总生存率分别为97.1%、6 9.4%、33.6%,高L3-SMI水平患者1、3、5年累积总生存率分别为98.4%、7 8.4%、52.3%，差异有统计学意义（x=31.718,P0.001）。术前低CONUT评分水平胃癌患者1、3、5年累积总生存率分别为98.8%8 0.2%、56.4%，高CONUT评分水平患者1、3、5年累积总生存率分别为96.7%6 7.6%、2 9.0%，差异有统计学意义（=48.240,P0.001）。单因素及多因素Cox回归分析结果显示：L3-SMI水平（高：HR=0.581,95%CI为0.459 0.7 35）及CONUT评分水平（高：HR=1.999,95%CI为1.57 4 2.539)是胃癌患者术后0 S的独立影响因素（P0.001)。根据多因素Cox回归分析构建胃癌患者术后0 S的列线图预测模型，其C指数为0.6 7 1(9 5%CI为0.6 39 0.703）,校准曲线显示预测曲线和真实曲线拟合度较好。结论：术前L3-SMI、CO NU T 评分是胃癌患者术后预后的独立影响因素,早期检测 L3-SMI、C O NU T 评分有助于预测胃癌患者术后预后情况，为临床决策提供参考和帮助。【关键词】骨骼肌质量指数；控制营养状况评分；胃癌；预后【中图分类号】R735.2【文章编号】16 7 2-4992-(2 0 2 3)2 1-398 5-0 7Predictive value of skeletal muscle mass index of the third lumbar spine and controllingnutritional status score for prognosis in patients with gastric cancerQIAN Xuewei,ZHU Qinqin,JIANG XiaohuiDepartment of Gastrointestinal Surgery,Afiliated Tumor Hospital,Nantong University,Jiangsu Nantong 226000,China.Abstract Objective:To investigate the prognostic value of skeletal muscle mass index of the third lumbar spine(L3-SMI)and controlling nutritional status(CONUT)scores in patients with gastric cancer after surgery.Methods:The clinical and survival data of 528 patients who underwent gastric cancer surgery in our hospital from January 2017to December 2020 were retrospectively analyzed.The optimal cut-off values of L3-SMI for patients of different gen-ders were determined by the receiver operating characteristic(ROC)curve.The clinical pathological characteristics ofpatients with different L3-SMI and CONUT scores were compared.The Kaplan-Meier method was used to analyzesurvival.Cox proportional hazards regression model was used to analyze the influencing factors for prognosis of patientswith gastric cancer by univariate and multivariated regression.Nomogram prediction models of overall survival(OS)of patients with gastric cancer were established by the R language software.The internal validation and accuracy【收稿日期】2023 05-11【修回日期】2023-07-11【基金项目】江苏省南通市科技计划指导性项目（编号：JCZ20133）【作者简介】钱雪蔚（1991一），女，江苏南通人，护师，主要从事胃肠肿瘤临床研究。【通信作者】江晓晖（196 8 一），男，江苏南通人，主任医师，硕士生导师，主要从事胃肠肿瘤临床研究。Ema i l:j i a n g x i a o h u i 0 518 16 3.c o m【文献标识码】AD0I:10.3969/j.issn.1672-4992.2023.21.015Modern Oncology 2023,31(21):3985-39913986of the nomograms were determined by the calculation of concordance index(C-index).Calibration curve was used toevaluate nomograms efficiency.Results:The best cut-off value of preoperative L3-SMI for predicting the prognosisof gastric cancer patients was 43.95 cm/m for male and 37.6 cm/m for female,and the area under the curve(AUC)values were 0.729(95%CI 0.672 0.785)and 0.680(95%CI 0.607 0.753).The 1-,3-and 5-yearcumulative overall survival rates were 97.1%,69.4%and 33.6%of patients with low L3-SMI level,and 98.4%,78.4%and 52.3%of patients with high L3-SMI level(x=31.718,P0.001).The 1-,3-and 5-year cumu-lative overall survival rates were 98.8%,80.2%and 56.4%of patients with low CONUT score,and 96.7%,67.6%and 29.0%with high CONUT score(x?=48.240,P0.001).The univariate and multivariate Cox regression analy-sis showed that L3-SMI level(high:HR=0.581,95%CI:0.459 0.735)and CONUT level(high:HR=1.999,95%ClI:1.574 2.539)were independent factors influencing the OS of patients with gastric cancer after surgery(P 4分为高CONUT组。1.3术后随访所有患者通过电话、查阅门诊或住院病历进行随访，了解生存状态。术后前2 年内每3个月随访1次,第35年每6个月随访1次,5年以后每年随访1次，随访检查项目包括血常规、肝肾功能、肿瘤标志物、胸腹部CT等。随访截止时间为2 0 2 2 年0 6 月30 日。总生存时间（overall survival,0S）定义为从手术至患者死亡或末次随访的时间。1.4统计学方法用SPSS26.0软件和R4.3.1软件进行统计分析，计量资料经Kolmogorov-Smirnov检验确认是否符合正态分布，正态性分布的资料以均数标准差(xs)表示,组间比较采用独立样本t检验，非正态性分布的资料采用中位数（M)和四分位数(P25,Prs)表示,组间比较采用非参数秩和检验。计数资料以频数和百分比表示，组间率的比较采用检验或Fisher精确概率法。采用Kaplan-Meier法绘制生存曲线,采用Log-Rank检验比较不同生存曲线间的差异,采用Cox回归模型分析影响胃癌患者预后的因素。使用R语言绘制列线图,采用Bootstrap法在原始数据集中随机抽取样本进行内部验证列线图的预测效能，用一致性指数（C指数）评价该模型的区分度,通过绘制校准曲线进行一致性验证。P65 years)Gender(female/male)BMI(25/25 kg/m)Drinking history(no/yes)Hypertension(no/yes)Diabetes(no/yes)ASA grade(I/I/II)Tumor site(fundusor cardia/corpus/antrumor pylorus)Tumor length(4/4 cm)Operation(open/laparoscope)Differentiation grade(well or moderate/poor or undifferentiated)TNM stage(1/I/II)T stage(T,/T,/T,/T)N stage(No/N,/N2/N3)Perineural invasion(no/yes)Vascular tumor thrombus(no/yes)CEA(5/5 ng/mL)Chemotherapy(no/yes)2.4不同 L3-SMICONUT水平胃癌患者生存曲线的比较本研究随访期内全组共32 3例死亡病例，中位生存时间为53个月，1、3、5年累积总生存率分别为97.7%、7 3.6%、41.9%。术前低L3-SMI水平胃癌患者中位生存时间为47(34,64）个月，1、3、5年累积总生存率分别为97.1%、69.4%、33.6%；术前高L3-SMI水平胃癌患者随访时间内中位生存时间缺失，1、3、5年累积总生存率分别为98.4%、78.4%、52.3%，两组患者生存情况比较，差异有统计学意义（x=31.718,P0.001），见图2。术前低CONUT评分水平胃癌患者中位生存时间缺失,1、3、5年累积总生存率分别为98.8%、8 0.2%、56.4%；术前高CONUT评分水平胃癌患者中位生存时间为45（32,6 3）个月，1、3、5年累积总生存率分别为96.7%、6 7.6%、2 9.0%，两组患者生存情况比较，差异有统计学意义（x=48.240,P 4分为高 CONUT组(2 7 6 例)。低L3-SMI组与高 L3-SMI 组患者在肿瘤最大径上差异有统计学意义(P0.05）,低 CONUT组与高 CONUT组患者在肿瘤分化程度及N分期上差异有统计学意义(P0.05）,见表1。1.00.80.6-0.40.20.01.00.80.60.40.2Specificity图1不同性别患者L3-SMI水平诊断术后总生存的ROC曲线Fig.1Receiver operating characteristic cures of L3-SMI in predictingthe prognosis of patients with different genders表 1不同 L3-SMI、CO NU T 水平胃癌患者临床病理特征的比较nPreoperative L3-SMILow level133/145103/175190/88146/132192/86240/3889/163/2639/93/146105/173171/107170/10846/119/11366/49/72/9187/53/76/62169/109175/103203/7535/243:3987.-FemaleMale0Preoperative CONUTPHigh level122/128109/141154/96135/115170/80218/3287/145/1832/83/135119/131146/104160/9044/112/9480/34/63/7390/42/54/64158/92167/83180/7042/208Fig.2Comparison of survival curves of patients with different preopera-tive L3-SMI levelsPLow levelHigh level0.0480.8262.350 0.1252.638 0.1040.1160.7330.069 0.7920.086 0.7691.0470.5920.2050.9035.2070.0220.5310.4660.456 0.5000.517 0.7725.1590.1613.605 0.3070.324 0.5690.855 0.3550.069 0.7931.8730.1711.000.750.500.250.00Numberofrisk(n)Lowlevel278270253Highlevel25024712Survivaltime(months)图2 术前不同L3-SMI水平胃癌患者生存曲线的比较132/120104/148165/87132/120178/74214/3880/152/2030/85/137104/148156/96179/7347/115/9072/33/68/79103/41/50/58158/94170/82187/6541/211P0.0011224Survival time(months)1981382351842436123/153108/168179/97149/127184/92244/3296/156/2441/91/144120/156161/115151/12543/116/11774/50/67/8574/54/80/68169/107172/104196/8036/240L3-SMIlevel+LoW+High+一-3648134483.222 0.0730.2510.6160.0220.8810.136 0.7120.9620.3271.3910.2380.7810.6770.9940.6080.2630.6080.7000.40314.9720.0002.618.0.2702.6510.44913.1830.0040.1200.7291.5260.2170.6740.4121.1010.294609710960:3988.1.00-0.75-0.500.250.00Numberofrisk(n)Lowlevel252249Highlevel27626812Survivaltime(months)图3术前不同CONUT评分水平胃癌患者生存曲线的比较Fig.3Comparison of survival curves of patients with different preopera-tiveCONUT levels2.5影响胃癌患者预后的单因素及多因素分析Cox回归模型对胃癌患者生存情况进行单因素分析结果Tab.2Univariate Cox regression analyses of factors affecting the prognosis of patients with gastric cancerVariableAge(65 years)Gender(female)BMI(25 kg/m)Drinking history(no)Hypertension(no)Diabetes(no)ASA grade(I)Tumor site(fundusor cardia)CorpusAntrumor pylorusTumor length(4 cm)Operation(open)Differentiation grade(well or moderate)TNM stage(I)T stage(T,)T2TT4N stage(No)NN2N3Perineural invasion(no)Vascular tumor thrombus(no)CEA(5 ng/mL)Chemotherapy(no)L3-SMI level(low)CONUT level(low)注：变量的括号内为对照。Note:The parentheses of the variables are for comparison.钱雪蔚，等L3骨骼肌质量指数与控制营养状况评分对胃癌患者术后预后的预测价值显示：患者年龄、ASA分级、肿瘤分化程度、TNM分期、T分CONUTlevel+Low-HighP0.0011224Survivaltime(months)2391972491852436期、N分期、脉管有癌栓、术后辅助化疗、L3-SMI及 CONUT评分与胃癌患者术后总生存时间有关(P0.05）,见表2。将这些危险因素结合临床实际情况进一步进行Cox回归模型进行多因素分析结果显示：ASA分级（I 级：HR=1.623，95%CI为1.0 91 2.415）、肿瘤分化程度（低、未分化：HR=1.416,95%CI为1.12 5 1.7 8 2)、脉管癌栓(有:HR=1.284,364815511748表2 影响胃癌患者预后的单因素Cox回归分析P0.2330.113-0.0150.113-0.0990.1180.0200.1120.0660.121-0.0440.167-0.0900.1210.4940.198-0.0640.179-0.0580.1690.0210.113-0.0540.1140.4250.1140.3640.1710.4990.1720.1440.1810.2730.1570.4450.1470.2800.1660.3340.1500.4810.1520.1430.1140.2820.1140.1910.1220.5910.1780.6430.1180.7840.11760129776095%CI为1.0 2 0 1.6 17)、TNM分期（II 期:HR=0.263,95%CI 为 0.12 1 0.57 2;期:HR=0.090,95%CI 为0.0 30 0.275)、T 分期（T:HR=2.470,95%CI为1.344 4.541;T4:HR=5.541,95%CI 为 2.490 12.32 9)、N 分期(N,:HR=1.948,95%CI 为 1.18 3 3.2 0 9;N2:HR=2.759,95%CI 为1.463 5.203;N,:HR=6.650,95%CI 为 2.8 8 7 15.316)、L3-SMI水平(高:HR=0.581,95%CI为0.459 0.7 35)及CONUT评分水平（高:HR=1.999,95%CI为1.57 4 2.539)是胃癌患者术后总生存时间的独立影响因素（P0.05）,见表3。Wald4.2690.0180.7100.0310.2940.0700.5566.2060.1270.1200.0360.22313.8424.5498.4060.6383.0149.1172.8584.9779.9871.5806.0862.44911.01029.94144.869HR0.0390.7920.8930.9850.3990.9060.8610.9810.5880.9370.7910.9570.4560.9140.0131.6400.7220.9380.7290.9430.8491.0220.6370.9480.0001.5300.0331.4390.0041.6480.4241.1550.0831.3140.0031.5610.0911.3240.0261.3970.0021.6180.2091.1540.0141.3260.1181.2110.0011.8050.0000.5260.0002.19195%CI0.635 0.9880.788 1.2300.719 1.1410.788 1.2200.739 1.1870.689 1.3280.721 1.1581.111 2.4190.661 1.3330.677 1.3140.819 1.2750.758 1.1851.223 1.9141.030 2.0101.176 2.3100.811 1.6460.965 1.7871.169 2.0840.956 1.8321.041 1.8741.200 2.1800.923 1.4421.060 1.6600.953 1.5391.274 2.5600.418 0.6621.741 2.756现代肿瘤医学2 0 2 3年11月第31卷第2 1期Tab.3 Multivariate Cox regression analyses of factors affecting the prognosis of patients with gastric cancerVariableAge(65 years)ASA stage(I)Differentiation grade(well or moderate)Perineural invasion(no)Vascular tumor thrombus(no)TNM stage(I)T stage(T,)T2TT4N stage(No)NN2N3Chemotherapy(no)L3-SMI level(low)CONUT level(low)注：变量的括号内为对照。Note:The parentheses of the variables are for comparison.2.6建立预测胃癌根治术后患者3、5年的OS列线图使用R4.3.1软件，纳入多因素Cox回归分析中影响OS的独立因素建立函数模型并绘制列线图,这些因素的每一个数字或类别在上面分数量表上的分数相加，对应在总分标上，向下画直线，与3、5年生存率坐标轴的交点表示每个时ScoreL3-SMIlevelCONUTlevelASAstageDifferentiationgradeVasculartumorthrombusTNMstageTstageNstageTotalpoints3-yearsurvivalrate5-yearsurvivalrateFig.4 A nomogram for predicting the OS of patients after radical gastrectomy at 3-and 5-year3讨论胃癌具有发病率高、死亡率高、预后差等特点，严重威胁人们的生命健康。由于胃癌早期症状较为隐匿，多数患者因病症就诊时已处于进展期，导致胃癌患者总体预后仍不尽人MODERN ONCOLOGY,Nov.2023,VOL.31,No.21表3影响胃癌患者预后的多因素Cox回归分析S-0.1990.1160.1080.1210.4840.2030.3480.1170.1150.1160.2500.1181.3360.397-2.4050.5690.3740.2680.9040.3111.7120.4080.6670.2551.0150.3241.8950.4260.3620.423-0.5430.1200.6930.122间点的估计生存时间或概率，见图4。其C指数为0.6 7 1（95%CI 为0.6 39 0.7 0 3）。校准曲线显示，0 S列线图预测模型对0 S的预测曲线与校准图中45对角线贴合度较好，提示该预测模型有较好的一致性，见图5。0102030405060708090100HighLowLowHighWellormoderatePoororundifferentiatedNoYesTN.50图4预测胃癌术后患者3、5年OS的列线图意,因此,早期评估胃癌患者的预后情况并识别预后较差的高危患者从而制定合理的治疗方案，对延长胃癌患者术后生存期具有重要的临床意义。本研究结果表明L3-SMI及CONUT评分是影响胃癌患者术后预后的独立影响因素。3989.WaldP2.9600.0850.7940.3735.7010.0178.7850.0030.9880.3204.5260.03311.3450.00117.8890.0001.9450.1638.4790.00417.6040.0006.8570.0099.8350.00219.8090.0000.7320.39220.5490.00032.2750.000工NN1001500.10.30.50.70.10.30.50.7HR0.8200.8971.6231.4161.1221.2840.2630.0901.4532.4705.5411.9482.7596.6501.4360.5811.999TN。2002500.995%CI0.654 1.0280.707 1.1391.091 2.4151.125 1.7820.894 1.4081.020 1.6170.121 0.5720.030 0.2750.8592.4581.344 4.5412.490 12.3291.183 3.2091.463 5.2032.88715.3160.627 3.2860.459 0.7351.574 2.5393003990Fig.5 Calibration chart of nomogram prediction model for predicting the 3-and 5-year survival rate of patients after radical gastrectomy术前营养状态是影响肿瘤患者接受治疗的耐受性及依从性的关键因素之一。营养不良不仅会影响肿瘤患者治疗方案的选择,降低对治疗的反应率,增加手术并发症和病死率,甚至对长期预后生存也有重要影响。有研究表明,约20%的肿瘤患者的死因并非肿瘤本身,而是死于营养不良及其并发症7 。临床中常见的营养评估方法有体重、BMI、握力等直接测量指标，血白蛋白、血红蛋白等血指标,以及营养风险筛查2 0 0 2、主观全面营养评价表等营养风险筛查评分工具，这些评估方法具有一定的局限性，有的甚至繁琐复杂、费时费力。肌肉减少症是一种营养不良状态,与肿瘤复发及不良预后有关,许多研究建议在临床实践中筛查患者是否患有肌肉减少症8-10 。研究发现第三腰椎肌肉水平与全身肌肉呈线性正相关,因此有学者提出利用CT技术自动测量L3-SMI,量化全身肌肉的质量状态,从而评估肿瘤患者的营养状态。L3-SMI已被多个研究证实与肿瘤患者预后生存相关12-14。本研究发现 L3-SMI 诊断预测胃癌瘤患者总生存时间的理想水平是男性为43.95cm/m、女性为37.6 cm/m，与ZHUANG等15 研究结果基本一致。术前骨骼肌状态被证实是胃癌患者术后的一项独立预后影响因素，但大多研究试验均限于6 5 岁的老年患者15-16 ,本研究纳人了18 8 0 岁的胃癌患者,结果显示L3-SMI水平仍是胃癌患者术后的独立预后因素。骨骼肌状态对肿瘤患者长期生存状态产生影响的原因考虑有：首先，骨骼肌细胞是人体最大的蛋白质库,可储存免疫系统所需的蛋白质及分泌免疫相关因子,如IL-6、I L-7、IL-15等,当体内骨骼肌减少时,这些可溶性细胞因子水平发生改变,失去正常的免疫调节功能致使抑制肿瘤功能受损17 。其次,骨骼肌作为外周吸收葡萄糖的主要组织之一，而当骨骼肌减少时，葡萄糖转运和利用功能受损导致体内胰岛素敏感性降低,最终出现胰岛素抵抗。胰岛素抵抗通过蛋白酶水解肌肉蛋白,释放支链氨基酸，从而影响肿瘤的发生发展18 。另外,骨骼肌细胞减少导致 IL-6分泌减少,会引起机体抗炎作用减弱，进而导致局部和全身慢性炎症的发生,致使肿瘤生长或复发19。因此,为了改善术前肌肉减少症状态肿瘤患者的长期生存,近年来提出一项新的围手术期管理策略一预康复2 0 。这一概念的兴起同样提示我们在关注肿瘤本身的同时，也要重视患者的心理、营养及运动干预2 1-2 CONUT评分是反映集体营养储备的血液评分指标,其评分越高则提示患者营养状态越差。血清ALB水平与营养不良的程度有关,体内肿瘤产生的肿瘤坏死因子（t u m o rnecrosis factor=，T NF）会导致肝细胞合成ALB 减钱雪蔚，等1.00.80.60.40.20.00.00.20.40.60.81.0Prediction of 3-year survival rate图5预测胃癌术后患者3、5年OS的列线图预测模型校准图L3骨骼肌质量指数与控制营养状况评分对胃癌患者术后预后的预测价值1.00.8-0.60.40.2-0.00.00.20.40.60.81.0Predictionof5-yearsurvivalrate少3。有研究表明血液胆固醇与肿瘤发生风险呈负相关，因为肿瘤组织会降低血液胆固醇水平及热量摄人2 4-2 5。而淋巴细胞计数反映机体免疫状态，当淋巴细胞计数较低时，肿瘤的预后较差2 6 。将这三种指标相结合形成CONUT评分，血清ALB的得分权重是胆固醇及淋巴细胞计数的两倍，从而可增加其评估肿瘤预后的准确性2 7 。本研究将 L3-SMI与 CONUT评分结合临床病理特征对胃癌患者术后的生存进行分析,结果显示 L3-SMI水平高的患者5年累积总生存率高于L3-SMI低的患者(52.3%vs33.6%,P0.001），C O NU T 评分低的患者5年累积总生存率高于CONUT评分高的患者（56.4%vs29.0%,P0.001）。进一步单因素及多因素 Cox回归分析证实 L3-SMI水平高(HR=0.581,95%CI 为 0.459 0.7 35)及 CONUT 评分高(HR=1.999,95%CI为1.57 4 2.539)是胃癌患者术后总生存时间的独立影响因素（P0.05）。结果表明L3-SMI、CONUT评分可作为一种新的预测胃癌患者预后指标,早期检测 L3-SMI、CO NU T 评分等相关因素并给予针对性干预，有望成为评估胃癌患者预后的工具。总之,L3-SMI、C O NU T 评分是胃癌患者术后总生存时间的独立影响因素，早期检测L3-SMI、C O NU T 评分可有助于评估胃癌患者术后预后情况。但由于本实验研究对象均为单中心病例,实验结果可能存在一定偏倚,期待未来有大样本、多中心的临床研究进一步证实L3SM I、C O NU T 评分的临床应用价值。1 CAO M,LI H,SUN D,et al.Cancer burden of major cancers in China:A need for sustainable actions J.Cancer Communications,2020,40(5):205-210.2STEWART OA,WU F,CHEN Y.The role of gastric microbiota ingastric cancerJ.Gut Microbes,2020,11(5):1220-1230.3MARIETTE C,DE BOTTON ML,PIESSEN G.Surgery in esopha-geal and gastric cancer patients:What is the role for nutrition sup-port in your daily practiceJJ.Ann Surg Oncol,2012,19(7):2128-2134.4 MARTIN L,BIRDSELL L,MACDONAD N,et al.Cancer cachexiain the age of obesity:Skeletal muscle depletion is a powerful prog-nostic factor,independent of body mass index J.J Clin Oncol,2013,31(12):1539 1547.5AHIKO Y,SHIDA D,HORIE T,et al.Controlling nutritional status(CONUT)score as a preoperative risk assessment index for olderpatients with colorectal cancer JJ.BMC Cancer,2019,19(1):946.6PRADO CM,LIEFFERS JR,MCCARGAR LJ,et al.Prevalence and【参考文献】现代肿瘤医学2 0 2 3年11月第31卷第2 1期clinical implications of sarcopenic obesity in patients with solidtumours of the respiratory and gastrointestinal tracts:A population-based studyJ.The Lancet Oncology,2008,9(7):629-635.7 WU BW,YIN T,CAO WX,et al.Clinical application of subjectiveglobal assessment in Chinese patients with gastrointestinal cancerJ.World J Gastroenterol,2009,15(28):3542-3549.8 PANJE CM,HONG L,HAYOZS S,et al.Skeletal muscle mass cor-relat