作者：付小菊，刘军，狄玉涵，周梦茹

单位：

223001 江苏省淮安市，徐州医科大学附属淮安医院 淮安市第二人民医院

Units：

Huaian Hospital Affiliated to Xuzhou Medical University/Huaian Second People's Hospital, Huaian 223001, China

关键词：

颅脑损伤；术后认知功能障碍；血糖变异性；影响因素分析；预测

Keywords：

Craniocerebral trauma; Postoperative cognitive dysfunction; Glucose variability; Root cause analysis;Forecasting

CLC：

R 651

DOI：

10.12114/j.issn.1008-5971.2021.00.139

Funds：

摘要：

背景　术后认知功能障碍（POCD）是颅脑损伤患者术后常见并发症，血糖波动与POCD 的发生有关， 而血糖变异性（GV）是衡量血糖波动程度的有效指标，故监测颅脑损伤患者术后GV 指标可能对预测POCD 的发生 具有重要意义。目的　探究GV 与颅脑损伤患者POCD 的关系。方法　选取2018 年5 月—2020 年10 月在徐州医科大 学附属淮安医院神经外科行手术治疗的颅脑损伤患者78 例为研究对象，根据患者术后15 d 是否发生POCD 将其分为 POCD 组25 例和非POCD 组53 例。比较两组患者一般资料、血糖指标〔空腹血糖（FPG）和糖化血红蛋白（HbA1c）、 24 h 平均血糖（24 h MBG）〕及GV 指标〔葡萄糖变异百分数（CV）、最大血糖波动幅度（LAGE）、24 h 内血糖波动 平均幅度（MAGE）及日间血糖平均绝对差（MODD）〕；24 h MBG、GV 指标与颅脑损伤伴POCD 患者HbA1c 的相关 性分析采用Pearson 相关分析；颅脑损伤患者发生POCD 的影响因素分析采用多因素Logistic 回归分析；绘制受试者工 作特征（ROC）曲线以评价CV、MAGE、MODD 对颅脑损伤患者发生POCD 的预测价值。结果　POCD 组患者HbA1c、 CV、LAGE、MAGE、MODD 均高于非POCD 组（P ＜ 0.05）。Pearson 相关分析结果显示，CV（r=0.519，P=0.008）、 LAGE（r=0.512，P=0.009）、MAGE（r=0.593，P=0.002） 及MODD（r=0.590，P=0.002） 与颅脑损伤伴POCD 患 者HbA1c 均呈正相关。多因素Logistic 回归分析结果显示，HbA1c〔OR=9.400，95%CI（1.444，61.200）〕、CV 〔OR=1.469，95%CI（1.033，2.088）〕、LAGE〔OR=9.043，95%CI（2.238，36.519）〕、MAGE〔OR=20.580，95%CI（1.887， 224.466）〕、MODD〔OR=17.163，95%CI（1.648，178.782）〕是颅脑损伤患者发生POCD 的独立影响因素（P ＜ 0.05）。 ROC 曲线分析结果显示，CV、LAGE、MAGE、MODD 预测颅脑损伤患者发生POCD 的曲线下面积（AUC）分别为0.780 〔95%CI（0.674，0.887）〕、0.740〔95%CI（0.623，0.856）〕、0.782〔95%CI（0.667，0.896）〕、0.766〔95%CI（0.656， 0.876）〕。结论　GV 指标CV、LAGE、MAGE、MODD 可能是颅脑损伤患者发生POCD 的影响因素，且四者均对颅脑损 伤患者发生POCD 具有一定预测价值，这对通过控制GV 预防颅脑损伤患者发生POCD 具有一定参考价值。

Abstract：

Background　Postoperative cognitive dysfunction (POCD) is a common complication in patients with craniocerebral injury. Glucose fluctuation is related to the occurrence of POCD, and glucose variability is an effective index to measure glucose fluctuation degree. Therefore, monitoring glucose variability in patients with craniocerebral injury may provide a theoretical basis for predicting the occurrence of POCD. Objective　To explore the correlation between glucose variability and POCD in patients with craniocerebral injury. Methods　A total of 78 patients with craniocerebral injury who underwent surgical treatment in Huaian Hospital Affiliated to Xuzhou Medical University from May 2018 to October 2020 were selected as the research objects, and patients were divided into POCD group (n=25) and non POCD group (n=53) according to whether POCD occurred at 15 days after operation. The general data, blood glucose indexes [fasting plasma glucose (FPG) and glycosylated hemoglobin A1c (HbA1c) , 24 h mean blood glucose (24 h MBG) ] and percentage coefficient of variation for glucose (CV) , largest amplitude of glycemic excursions (LAGE) , mean amplitude of glycemic excursions (MAGE) and means of daily differences (MODD) were compared between the two groups; Pearson correlation analysis was used to analyze the correlation between 24 h MBG, GV indexes and HbA1c in patients with craniocerebral injury complicated with POCD; the multivariate Logistic regression analysis was used to analyze the influencing factors of POCD in patients with craniocerebral injury; and receiver operating characteristic (ROC) curve was drawn to evaluate the predictive value of CV, MAGE, MODD on POCD in patients with craniocerebral injury. Results　Compared with the non POCD group, the HbA1c, CV, LAGE, MAGE and MODD in the POCD group were significantly increased (P

ReferenceList：