2型糖尿病血糖变异性对腰椎椎间融合术后手术部位感染的影响

Effects of glycemic variability in type 2 diabetes on surgical site infection after lumbar interbody fusion

中文关键词:  2型糖尿病  血糖变异性  经椎间孔腰椎椎间融合术  手术部位感染

英文关键词:Type 2 diabetes mellitus  Glycemic variability  Transforaminal lumbar interbody fusion  Surgical site infection

中文摘要:

  【摘要】 目的：分析2型糖尿病患者围手术期血糖变异性与经椎间孔腰椎椎间融合术（transforaminal lumbar interbody fusion，TLIF）术后手术部位感染（surgical site infection，SSI）的相关性。方法：回顾分析 2018年1月~2021年4月305例在东南大学附属中大医院脊柱外科接受TLIF的合并2型糖尿病的腰椎退行性疾病（lumbar degenerative diseases，LDD）患者，其中男性133例，女性172例，平均年龄67.6±9.3岁。收集所有患者的病历资料，一般资料包括性别、年龄、体质指数（body mass index，BMI）（是否≥25kg/m2）、糖尿病病程、是否合并高血压、是否合并冠心病、入院空腹血糖、术前糖化血红蛋白（glycosylated hemoglobin A1c，HbA1c）、术前及术后平均空腹血糖（mean fasting blood glucose，MFBG）、术前降糖方案（①口服降糖药；②皮下注射胰岛素；③联合用药：口服降糖药+皮下注射胰岛素；④饮食疗法）。手术相关资料包括术中出血量、术中输血量、手术时间、手术节段（是否≥2个）、术后引流时间、术后引流量、切口长度。血糖变异性（glycemic variability，GV）监测指标包括术前及术后的空腹血糖水平的标准差（standard deviation of blood glucose，SDBG）、空腹血糖变异系数（coefficient of variation，CV）、空腹血糖最大变异幅度（largest amplitude of glycemic excursions，LAGE）、日间血糖平均绝对差（mean of daily differences，MODD）。根据SSI诊断标准确定术后SSI病例，将患者分为感染组与非感染组。对比两组患者术前及术后的血糖变异性指标，即SDBG、CV、LAGE、MODD等进行分析，同时通过相关性分析和受试者工作特征曲线（receiver operating characteristic curve，ROC）探讨围手术期血糖变异性参数和患者TLIF术后SSI的关系及其预测价值。结果：305例患者中发生SSI者51例，单因素分析显示糖尿病病程、术前及术后的MFBG、SDBG、CV、LAGE、MODD、术前降糖方案、手术节段、术后引流时间、切口长度等因素两组间比较差异有统计学意义（P<0.05）。而在性别、年龄、BMI、高血压、冠心病、入院空腹血糖、术前HbA1c、手术时间、术中出血量、术中输血量、术后引流量等因素比较差异无统计学意义（P＞0.05）。术前SDBG、CV、LAGE、MODD曲线下面积（areas under the curve，AUC）分别为0.840、0.813、0.851、0.680，cut-off值分别是0.89mmol/L、13.69%、2.25mmol/L、0.92mmol/L。术后SDBG、CV、LAGE、MODD曲线下面积分别为0.697、0.672、0.693、0.698，cut-off值分别是1.72mmol/L、16.09%、3.95mmol/L、1.59mmol/L。岭回归多因素分析结果显示术前MFBG、SDBG、CV、LAGE、MODD及术后SDBG、CV、LAGE、MODD大以及术后引流时间长、多手术节段均是2型糖尿病患者发生TLIF术后SSI的独立危险因素（P<0.05），而感染组与非感染组术后平均空腹血糖差异无统计学意义（P>0.05）。结论：2型糖尿病患者围手术期血糖变异性与TLIF术后SSI的发生密切相关，减小血糖变异性可能有助于降低术后SSI发生率。

英文摘要:

  【Abstract】 Objectives: To analyze the correlation between perioperative glycemic variability and surgical site infection(SSI) following transforaminal lumbar interbody fusion(TLIF) in patients with type 2 diabetes mellitus. Methods: This study retrospectively analyzed 305 patients with lumbar degenerative diseases(LDD) and type 2 diabetes who underwent TLIF in the Spinal Surgery Department of Zhongda Hospital Affiliated to Southeast University from January 2018 to April 2021. There were 133 males and 172 females with an average age of 67.6±9.3 years. The medical records of all patients were collected, and postoperative infection cases were determined according to the diagnostic criteria of surgical site infection. General information included gender, age, whether combined with body mass index(BMI)≥25kg/m2, duration of diabetes mellitus, whether combined with hypertension, whether combined with coronary heart disease, fasting blood glucose on admission, preoperative glycosylated hemoglobin A1c(HbA1c), preoperative and postoperative mean fasting blood glucose(MFBG), and preoperative hypoglycemic scheme(①oral hypoglycemic drugs; ②subcutaneous insulin injection; ③combined medication: oral hypoglycemic drugs + subcutaneous insulin injection; ④diet therapy). Surgery-related data included intraoperative blood loss, intraoperative blood transfusion, duration of operation, the number of operative levels≥(2 or not), postoperative drainage time, postoperative drainage volume, and length of the incision. Glycemic variability monitoring indicators included standard deviation of blood glucose(SDBG), coefficient of variation(CV), largest amplitude of glycemic excursions(LAGE), and mean of daily differences (MODD). Patients were divided into the infection group and non-infection. Preoperative and postoperative glycemic variability indexes of the above 2 groups were compared, including SDBG, CV, LAGE, and MODD. The correlation analysis and receiver operating characteristic(ROC) curve were used to investigate the relationship between perioperative glycemic variability and postoperative SSI and its predictive value. Results: A total of 51 patients out of the 305 patients occurred infection. Univariate analysis showed that there were significant differences between the two groups in the duration of diabetes mellitus, preoperative and postoperative MFBG, SDBG, CV, LAGE, MODD, preoperative hypoglycemic regimen, the number of operative levels, postoperative drainage time, and the length of incision(P<0.05). However, there were no significant differences in gender, age, BMI, hypertension, coronary heart disease, fasting blood glucose on admission, preoperative HbA1c, operation time, intraoperative blood loss, intraoperative blood transfusion, and postoperative drainage volume(P>0.05). ROC analysis showed that areas under the curve(AUC) of preoperative SDBG, CV, LAGE, and MODD were 0.840, 0.813, 0.851, and 0.680, and cut-off values were 0.89mmol/L, 13.69%, 2.25mmol/L, and 0.92mmol/L respectively. The AUC of postoperative SDBG, CV, LAGE, and MODD were 0.697, 0.672, 0.693, 0.698, and cut-off values were 1.72mmol/L, 16.09%, 3.95mmol/L, and 1.59mmol/L respectively. The ridge regression results showed that the high preoperative MFBG, SDBG, CV, LAGE, and MODD, and postoperative SDBG, CV, LAGE, MODD, prolonged postoperative drainage duration, and multiple operative segments were found to be the independent risk factors for postoperative SSI following TLIF in patients with type 2 diabetes mellitus (P<0.05), while there was no significant difference in postoperative MFBG between the infection group and non-infection group(P>0.05). Conclusions: Perioperative glycemic variability in patients with type 2 diabetes is closely related to the occurrence of postoperative SSI following TLIF. Reducing blood glucose variability may be beneficial to reduce the incidence of SSI after surgery.