肺癌合并糖尿病患者血糖波动对预后的影响

Effect of blood glucose fluctuation on the prognosis of patients with diabetes and lung cancer

目的探讨肺癌合并糖尿病患者血糖波动对预后的影响。方法选取2019年3月至2020年5月间宝鸡市中心医院收治的92例肺癌合并糖尿病患者,依据180天死亡事件分为存活组65例和死亡组27例。观察两组患者血糖波动情况,分析血糖波动指标与患者预后的相关性。结果两组患者血糖波动频数比较,差异无统计学意义(P> 0.05);死亡组患者初始血糖(BGad)、24h平均血糖(BGm)、标准差(BGsd)、血糖变异系数(BGcV)和血糖不稳定指数(BGI)水平均高于生存组,差异均有统计学意义(均P <0.05)。Pearon相关分析显示,患者BGad、BGm、BGsd、BGcV和BGI水平均与预后呈正相关(均P <0.05)。Logistic回归分析显示,BGad、BGm、BGsd、BGcV和BGI水平过高,均为肺癌合并糖尿病患者死亡的独立危险因素(均P <0.05)。将BGad、BGm、BGsd、BGcV和BGI指标纳入二元Logistic回归模型行统计分析,得到五项联合指标。再建立各自ROC曲线。分析结果显示,BGad、BGm、BGsd、BGcV、BGI及五项联合指标预测患者预后ROC曲线下面积AUC分别为0.665、0.759、0.773、0.980、0.924和0.993。可知,五项联合指标预测患者预后的价值高于单项指标,其敏感度、特异度分别为92.6%和100.0%。结论血糖波动与肺癌合并糖尿病患者预后有密切相关性,为患者死亡的独立危险因素,可作为患者预后的预测指标。

Objective To explore the effect blood glucose fluctuation on the prognosis of patients with diabetes and lung cancer. Methods From March 2019 to May 2020,a total of 92 patients with diabetes and lung cancer were selected at Baoji Central Hospital. The patients were divided into a survived group( 65 patients) and a died group( 27 patients) based on 180 d death events. Blood glucose fluctuations were observed in the two groups and the association between the blood glucose fluctuation and the prognosis of patients with diabetes and lung cancer were analyzed. Results There was no significant difference in blood glucose fluctuation between the two groups( P > 0. 05). The levels of blood glucose on admission( BGad),mean blood glucose( BGm),standard deviation blood glucose( BGsd),coefficient of blood glucose variability( BGcv) and blood glucose instability index( BGI) were significantly higher in the died group than in the survived group( all P < 0. 05). Pearson correlation analysis showed that the levels of BGad,BGm,BGsd,BGcV and BGI were positively associated with prognosis( all P < 0. 05). Logistic regression analysis showed that excessive levels of BGad,BGm,BGsd,BGcV and BGI were independent risk factors for death in patients with diabetes and lung cancer( all P < 0. 05). Binary Logistic regression model was used to analyze BGad,BGm,BGsd,BGcV and BGI resulting in the five combination indicators. Receiver operating characteristics( ROC) was created for each of them and the analysis showed that area under the ROC curve( AUC) was 0. 665,0. 759,0. 773,0. 980,0. 924 and 0. 993,respectively for BGad,BGm,BGsd,BGcV,BGI and the five combination indicators. It revealed that the value of the five combination indicators is higher than that of each single indicator for predicting the prognosis of patients with diabetes and lung cancer.Sensitivity and specificity was 92. 6% and 100. 0%,respectively. Conclusion Blood glucose fluctuation is closely associated with the prognosis of patients with diabetes and lung cancer. It is an independent risk factor for death and can be used as a predictor for the prognosis of patients with diabetes and lung cancer.