目的 分析第三腰椎旁肌肉指数( L3 -SMI) 与晚期非小细胞肺癌(NSCLC) 营养状态之间的关系,以进一步明确 L3 -SMI 是否可以作为评估 NSCLC 营养不良的有效预测参数。 方法 选择 2016 年 1 月至 2020 年 12 月间于阜宁县人民医院行 放、化疗的 100 例Ⅲ~Ⅳ期 NSCLC 为研究对象,并根据患者主观整体评估(PG-SGA)将所有患者分为重度营养不良组与非重 度营养不良组。 采用单、多因素 Logistic 回归分析确定各评价参数与重度营养不良的相关性,并通过受试者操作特征(ROC) 曲线比较不同评价参数对 NSCLC 重度营养不良的预测价值。 结果 PG-SGA 量表发现严重营养不良在进展期或晚期 NSCLC 中的发生率高达 48. 0%。 与非重度营养不良组相比,重度营养不良组年龄更大[( 65. 4 ± 10. 0) 岁比 ( 61. 2 ± 6. 9) 岁, t = 2. 482, P= 0. 015)],控制营养状态评分更高[3(1,5)分比 2(1,4)分, t = 2. 238, P = 0. 027],而体质指数[(21. 1±3. 2) kg / m 2 比(25. 2±3. 8) kg / m 2 , t = - 5. 840, P< 0. 001)]、血清白蛋白水平[( 33. 8± 4. 7) g / L 比( 36. 4± 3. 9) g / L, t = - 2. 970, P = 0. 004)]、预后营养指数[( 41. 5 ± 5. 3) 比( 44. 7 ± 5. 0), t = - 3. 122, P = 0. 002)] 及 L3 - SMI 更低[( 36. 0 ± 8. 9) 比( 53. 1 ± 12. 3),t = -8. 103, P<0. 001)]。 多因素 Logistic 回归分析证实血清低白蛋白水平(OR = 0. 791, 95%CI = 0. 678 ~ 0. 922,P = 0. 003)和低 L3 -SMI(OR= 0. 826, 95%CI = 0. 765~ 0. 892,P<0. 001)是严重营养不良的危险因素。 ROC 曲线展示 L3 -SMI 对 NSCLC 的营养状态具有良好预测效能(AUC 值: 0. 888, 95% CI = 0. 822 ~ 0. 955),优于其他血液学参数。 结论 L3 -SMI 与PG-SGA 具有高度一致性,是一项筛查晚期 NSCLC 营养不良风险的有效指标。

Objective This study aimed to analyze the correlation between the third lumbar skeletal muscle index L3 -SMI and nutritional status and to clarify whether L3 -SMI could be used as an effective predictor of malnutrition for advanced non-small cell lung cancer NSCLC patients. Method A total of 100 clinical stage Ⅲ - Ⅳ NSCLC patients who received radiotherapy and / or chemotherapy in Funing People's Hospital from Jan 2016 to Dec 2020 were enrolled in this analysis and all patients were divided into severe malnutrition and non-severe malnutrition groups based on the patient-generated subjective global assessment PG-SGA . The correlations between different assessment parameters and severe malnutrition risk were determined using the univariate and multivariate Logistic regression analysis and the receiver operation characteristics ROC curves were plotted to compare these predictive values for severe malnutrition risk in advanced NSCLC patients. Result Using the PG-SGA scale we found that frequency of severe malnutrition in advanced NSCLC patients could be as high as 48. 0% . Compared with the non -severe malnutrition group patients with severe malnutrition were older 65. 4± 10. 0 vs 61. 2 ± 6. 9 years old t = 2. 482 P = 0. 015 and had a higher score of Controlling nutritional status 3 1 5 vs 2 1 4 t = 2. 238 P = 0. 027 lower body mass index 21. 1±3. 2 kg / m 2 vs 25. 2±3. 8 kg / m 2 t = -5. 840 P<0. 001 serum albumin level 33. 8±4. 7 vs 36. 4±3. 9 g / L t = -2. 970 P= 0. 004 the prognostic nutritional index 41. 5±5. 3 vs 44. 7±5. 0 t = -3. 122 P= 0. 002 and L3 -SMI 36. 0±8. 9 vs 53. 1±12. 3 t = -8. 103 P<0. 001 . The multivariate Logistic regression analysis demonstrated that low serum albumin level OR = 0. 791 95%CI = 0. 678- 0. 922 P = 0. 003 and L3 -SMI OR= 0. 826 95%CI = 0. 765-0. 892 P<0. 001 were independently correlated with severe malnutrition risk in advanced NSCLC patients. In addition ROC curves showed that L3 -SMI had an excellent predictive performance with an AUC value of 0. 888 95%CI = 0. 822 - 0. 955 which was higher than other hematological parameters. Conclusion L3 -SMI was significantly correlated with PG-SGA scale and it was an useful tool of nutrition risk screening for advanced NSCLC patients.