陳慧敏 賈洪峰 江婷婷 賈耀輝
摘要:目的 分析術(shù)中血糖波動和術(shù)后胰島素抵抗(IR)對老年全麻胸腔鏡肺癌根治術(shù)患者術(shù)后認知功能障礙(POCD)的影響。方法 收集352例老年全麻胸腔鏡肺癌根治術(shù)患者,分為POCD組(84例)和非POCD組(268例),利用傾向性評分匹配法(PSM)均衡2組間協(xié)變量,84例匹配成功,比較PSM前后2組資料差異。PSM后,繪制血糖波動幅度預測POCD的受試者工作特征(ROC)曲線,根據(jù)截斷值分為低水平血糖波動組(97例)和高水平血糖波動組(71例),并根據(jù)術(shù)后是否發(fā)生IR分為IR組(53例)和非IR組(115例),比較組間POCD發(fā)生率。采用Logistic回歸分析POCD的影響因素。結(jié)果 PSM前,POCD組的年齡、血糖波動幅度、IR比例、手術(shù)時間、麻醉時間、麻醉維持期間丙泊酚用量、瑞芬太尼用量和舒芬太尼用量大于非POCD組(P<0.05);PSM后,POCD組血糖波動幅度和IR比例大于非PCOD組(P<0.05)。PSM后,高水平血糖波動組和IR組POCD發(fā)生率分別高于低水平血糖波動組和非IR組(P<0.05)。Logistic回歸顯示,術(shù)中血糖波動幅度大(OR=9.140,95%CI:4.338~19.257)和術(shù)后發(fā)生IR(OR=4.034,95%CI:1.163~13.991)是POCD的危險因素。結(jié)論 術(shù)中血糖波動幅度大和術(shù)后發(fā)生IR的老年全麻胸腔鏡肺癌根治術(shù)患者POCD發(fā)生風險升高。
關(guān)鍵詞:血糖;胰島素抵抗;胸腔鏡檢查;麻醉,全身;術(shù)后認知并發(fā)癥;認知功能障礙;老年人;肺癌根治術(shù)
中圖分類號:R614.2文獻標志碼:ADOI:10.11958/20230390
Effects of intraoperative blood glucose fluctuation and postoperative insulin resistance on cognitive dyfunction in elderly patients after thoracoscopic radical resection of?lung cancer under general anesthesia
Abstract: Objective To analyze the influence of intraoperative blood glucose fluctuation and postoperative insulin resistance (IR) on postoperative cognitive dyfunction (POCD) in elderly patients undergoing thoracoscopic radical resection of lung cancer under general anesthesia. Methods A total of 352 elderly patients undergoing thoracoscopic radical resection of lung cancer under general anesthesia were collected and divided into the POCD group (n=84) and the non-POCD group (n=268). The covariates between the two groups were balanced by propensity score matching method (PSM). Eighty-four cases in each group were successfully matched. The data between the two groups before and after PSM were compared. After PSM, receiver operating characteristic (ROC) curve of blood glucose fluctuation amplitude for predicting POCD was drawn, and patients were divided into the low-level blood glucose fluctuation group (n=97) and the high-level blood glucose fluctuation group (n=71) according to the cut-off value. According to the existence of postoperative IR, patients were divided into the IR group (n=53) and the non-IR group (n=115). Then, incidences of POCD between groups were compared. Logistic regression was used to analyze the influencing factors of POCD. Results Before PSM, the POCD group had older age, higher blood glucose fluctuation amplitude, IR ratio, operation time, anesthesia time, propofol dosage, remifentanil dosage and sufentanil dosage in anesthesia maintenance period than those in the non-POCD group (P<0.05). The POCD group had higher blood glucose fluctuation amplitude and IR ratio than those in the non-POCD group after PSM (P<0.05). After PSM, the incidences of POCD in the high-level blood glucose fluctuation group and the IR group were higher than those in the low-level blood glucose fluctuation group and the non-IR group (P<0.05). Logistic regression analysis showed that higher intraoperative blood glucose fluctuation (OR=9.140, 95%CI: 4.338-19.257) and postoperative IR (OR=4.034, 95%CI: 1.163-13.991) were risk factors of POCD. Conclusion The risk of POCD in elderly patients undergoing thoracoscopic radical lung cancer surgery under general anesthesia is increased in patients with higher intraoperative blood glucose fluctuation and postoperative IR.
Key words: blood glucose; insulin resistance; thoracoscopy; anesthesia, general; postoperative cognitive complications; cognitive dysfunction; aged; radical lung cancer surgery
全麻胸腔鏡肺癌根治術(shù)是肺癌的重要治療手段。術(shù)后認知功能障礙(postoperative cognitive dysfunction,POCD)是全麻術(shù)后常見并發(fā)癥,表現(xiàn)為意識混亂、記憶受損、注意力減退等癥狀,其發(fā)生率較高,嚴重影響患者生活質(zhì)量[1-2]。POCD的發(fā)病機制尚不明確,既往研究多認為與手術(shù)時間、麻醉時間、麻醉藥物、年齡等有關(guān)[3-4],但這些因素不能完全解釋POCD的發(fā)生,有必要進一步探討。手術(shù)患者在術(shù)中會表現(xiàn)出一定應激反應,分泌大量胰島素拮抗激素,導致血糖升高,出現(xiàn)血糖波動[5]。圍手術(shù)期胰島素抵抗(insulin resistance,IR)是機體對胰島素敏感性和反應性降低的一種病理狀態(tài)[6]。血糖波動、IR與認知功能相關(guān)[7-8]。但術(shù)中血糖波動和術(shù)后IR對POCD的影響尚不清楚。本研究探討了術(shù)中血糖波動和術(shù)后IR對老年全麻胸腔鏡肺癌根治術(shù)患者POCD的影響,旨在為POCD的預防提供參考依據(jù)。
1 對象與方法
1.1 研究對象 選取2021年1月—2022年12月于西安交通大學醫(yī)學部附屬三二〇一醫(yī)院進行手術(shù)治療的352例老年肺癌患者作為研究對象。納入標準:(1)病理學確診為肺癌,接受胸腔鏡肺癌根治術(shù)治療。(2)年齡≥60歲,受教育程度中學及以上。(3)術(shù)前1 d行簡易智力狀況檢查(mini-mental state examination,MMSE)量表評估提示該評分在正常范圍,即27~30分。(4)麻醉方式為全麻,全麻藥物給藥方法一致;美國麻醉醫(yī)師協(xié)會(American Society of Anesthesiologists,ASA)分級Ⅱ或Ⅲ級。(5)患者能夠正常交流,配合量表評估。(6)心肺、肝腎功能無明顯異常。排除標準:(1)有腦部外傷史、腦部手術(shù)史或腹部手術(shù)史。(2)腦血管疾病后遺癥。(3)精神疾病。(4)長期服用精神類藥物或酗酒,長期服用激素等導致血糖升高藥物。(5)凝血功能障礙、自身免疫系統(tǒng)疾病、糖尿病。(6)手術(shù)或者麻醉禁忌證。本研究獲得所有患者及家屬知情同意,經(jīng)醫(yī)院倫理委員會審批通過(批準編號2020016)。
1.2 麻醉方法和患者自控鎮(zhèn)痛(patient controlled analgesia,PCA) 麻醉誘導方案:先后靜脈注射丙泊酚2.5 mg/kg、瑞芬太尼0.5 μg/kg、順式阿曲庫銨0.025 mg/kg,誘導結(jié)束后氣管插管維持通氣。麻醉維持方案:持續(xù)泵入丙泊酚5 mg/(kg·h),瑞芬太尼0.5 μg(/kg·h),間斷泵入舒芬太尼0.2 μg/kg,手術(shù)結(jié)束后停止應用上述麻醉藥物。麻醉過程中未應用其他藥物。術(shù)后常規(guī)給予PCA鎮(zhèn)痛。
1.3 術(shù)中血糖波動和術(shù)后IR監(jiān)測 于手術(shù)開始時、術(shù)中每30 min、手術(shù)結(jié)束時監(jiān)測血糖,取血糖監(jiān)測值的最大值和最小值,二者之差為血糖波動幅度。于術(shù)后1~5 d抽取患者空腹靜脈血,檢測血糖和胰島素水平。根據(jù)胰島素抵抗穩(wěn)態(tài)模型(homeostasis model assessment of inusilin resistance,HOMA-IR)判斷是否存在IR。HOMA-IR=空腹血糖(mmol/L)×空腹胰島素(mU/L)/22.5,HOMA-IR≥1.73為存在IR[9]。
1.4 POCD評判標準 術(shù)前1 d及術(shù)后1周,采用MMSE量表評估患者認知功能,術(shù)后1周MMSE評分較術(shù)前1 d降低超過2分則判定為發(fā)生POCD[10]。根據(jù)是否發(fā)生POCD將患者分為POCD組84例和非POCD組(非POCD組1)268例。
1.5 收集資料 收集患者性別、年齡、體質(zhì)量指數(shù)(BMI)、吸煙、飲酒、合并高血壓、合并高脂血癥、ASA分級、術(shù)前1 d MMSE評分、術(shù)中血糖波動幅度、IR、手術(shù)時間、麻醉時間、麻醉維持期間丙泊酚用量、瑞芬太尼用量、舒芬太尼用量等臨床資料。
1.6 單因素分析 使用傾向性評分匹配法(propensity score matching,PSM)1∶1匹配POCD組和非POCD組臨床資料相近的患者,需要匹配的變量為年齡、手術(shù)時間、麻醉時間、麻醉維持期間丙泊酚用量、瑞芬太尼用量、舒芬太尼用量。PSM匹配后,將匹配成功的84例非COPD患者作為非COPD組2。比較PSM前后POCD組和非POCD組患者臨床資料差異。
1.7 POCD發(fā)生率比較 繪制血糖波動幅度預測POCD發(fā)生的受試者工作特征(ROC)曲線,根據(jù)血糖波動幅度預測POCD發(fā)生的最佳截斷值將患者分為低水平血糖波動組(血糖波動<3.15 mmol/L,97例)和高水平血糖波動組(血糖波動≥3.15 mmol/L,71例),并根據(jù)術(shù)后是否發(fā)生IR將患者分為IR組(53例)和非IR組(115例),比較低、高水平血糖波動組間及IR組、非IR組間的POCD發(fā)生率。
1.8 多因素二元Logistic回歸分析 將PSM后POCD組和非POCD組差異有統(tǒng)計學意義的因素納入二元Logistic回歸模型,分析影響老年全麻胸腔鏡肺癌根治術(shù)患者發(fā)生POCD的危險因素。
1.9 統(tǒng)計學方法 采用SPSS 23.0軟件和R軟件(R4.3.1)進行分析。計數(shù)資料以例(%)表示,組間比較采用χ2檢驗。計量資料以均數(shù)±標準差([[x] ±s
])表示,組間比較采用t檢驗。采用R軟件進行PSM分析,1∶1匹配POCD組和非POCD組需要平衡的變量,選擇最鄰近匹配。采用ROC曲線分析血糖波動幅度對POCD發(fā)生的預測價值。采用多因素二元Logistic回歸分析發(fā)生POCD的影響因素。P<0.05為差異有統(tǒng)計學意義。
2 結(jié)果
2.1 PSM前后患者POCD的單因素分析 PSM前,POCD組年齡、IR比例、血糖波動幅度、手術(shù)時間、麻醉時間、麻醉維持期間丙泊酚用量、瑞芬太尼用量和舒芬太尼用量大于非POCD組(P<0.05)。PSM后,POCD組僅IR比例和血糖波動幅度比例大于非POCD組(P<0.05)。見表1。
2.2 術(shù)中血糖波動幅度對患者發(fā)生POCD的預測價值 PSM后,血糖波動幅度預測患者POCD的AUC為0.933,95%CI為0.894~0.972(P<0.01),截斷值為3.15 mmol/L,敏感度為82.1%,特異度為97.6%,見圖1。
2.3 血糖波動幅度和IR對患者POCD的影響 高水平血糖波動組POCD發(fā)生率(97.2%,69例)高于低水平血糖波動組(15.5%,15例,χ2=109.504,P<0.01)。IR組POCD發(fā)生率(69.8%,37例)高于非IR組(40.9%,47例,χ2=12.156,P<0.01)。
2.4 患者發(fā)生POCD的影響因素分析 以是否發(fā)生POCD為因變量(是=1,否=0),以血糖波動幅度、IR(是=1,否=0)為自變量,進行多因素Logistic回歸分析。結(jié)果顯示,術(shù)中血糖波動幅度大和術(shù)后發(fā)生IR是老年全麻胸腔鏡肺癌根治術(shù)患者POCD的危險因素(P<0.05),見表2。
3 討論
3.1 全麻胸腔鏡肺癌根治術(shù)的老年患者具有較高的POCD發(fā)生率 本研究中,接受全麻胸腔鏡肺癌根治術(shù)的老年患者術(shù)后1周時的POCD發(fā)生率為23.9%,與既往數(shù)據(jù)中非心臟手術(shù)患者POCD發(fā)生率在17.1%~41.4%的結(jié)果類似[11],而低于蘇容容等[12]研究中胸腔鏡根治術(shù)患者37.0%的POCD發(fā)生率。分析原因,一方面可能與手術(shù)技術(shù)、圍手術(shù)期管理的進步有關(guān);另一方面可能與術(shù)后常規(guī)應用PCA鎮(zhèn)痛減少疼痛刺激傳入、降低氧化應激和中樞炎性反應有關(guān)[13]。但是,本研究中老年全麻胸腔鏡肺癌根治術(shù)患者POCD的發(fā)生率仍然較高,主要與老年人生理功能逐漸減退,對手術(shù)的應激能力和藥物的代謝能力下降,腦細胞代謝障礙風險明顯增加等因素有關(guān)[14]。因此,識別老年全麻胸腔鏡肺癌根治術(shù)患者POCD的危險因素并進行針對性干預具有重要臨床意義。
3.2 術(shù)中血糖波動幅度大是POCD的危險因素 對于手術(shù)患者,一方面手術(shù)創(chuàng)傷會誘發(fā)機體分泌兒茶酚胺、皮質(zhì)醇等多種胰島素拮抗因子,促使血糖升高;另一方面長時間禁食可能會引起患者低血糖,可增加圍手術(shù)期并發(fā)癥發(fā)生風險[15]。魯國等[16]在急性腦梗死伴糖尿病患者認知功能障礙相關(guān)研究中發(fā)現(xiàn),認知功能障礙組患者的血糖波動幅度、血糖變異率等血糖變異指標高于認知功能正常組患者。Lim等[17]研究認為血糖變化在預測卒中后認知功能障礙方面具有重要價值。以上研究均提示血糖波動與認知功能有關(guān)。但關(guān)于術(shù)中血糖波動對POCD影響的研究較少。本研究顯示,術(shù)中血糖波動幅度對POCD發(fā)生具有較好的預測價值,高水平血糖波動組患者POCD發(fā)生率高于低水平血糖波動組,且術(shù)中血糖波動幅度大是老年全麻胸腔鏡肺癌根治術(shù)患者發(fā)生POCD的危險因素。分析原因,機體的抗氧化平衡機制在高血糖和低血糖的狀態(tài)下會發(fā)生改變,導致氧化應激和炎癥反應加重,從而導致神經(jīng)炎癥和認知功能障礙[18-19]。因此,在手術(shù)過程中,應當密切監(jiān)測血糖,采取合理的控制血糖方案,保持血糖平穩(wěn),避免血糖劇烈波動。
3.3 術(shù)后發(fā)生IR是老年全麻胸腔鏡肺癌根治術(shù)患者POCD的危險因素 圍手術(shù)期IR是指手術(shù)患者在應激下表現(xiàn)出的對胰島素的敏感性和反應性降低的一種病理狀態(tài),嚴重影響患者的術(shù)后恢復[20]。但術(shù)后發(fā)生IR是否會增加POCD的發(fā)生風險尚不清楚。本研究顯示,IR組患者POCD發(fā)生率高于非IR組,且術(shù)后發(fā)生IR是老年全麻胸腔鏡肺癌根治術(shù)患者發(fā)生POCD的危險因素。這可能與以下機制有關(guān):(1)IR可引起血管內(nèi)皮功能障礙,加重機體的炎癥反應和氧化應激[21-22]。(2)IR可引起腦組織低灌注,導致腦組織缺血缺氧[23]。(3)IR導致的外周高胰島素血癥會引發(fā)海馬體氧化應激反應和神經(jīng)炎癥,導致認知功能障礙的發(fā)生[24]。因此,應當關(guān)注圍手術(shù)期IR,從減輕IR程度的角度達到預防POCD的目的。
綜上所述,老年全麻胸腔鏡肺癌根治術(shù)患者具有較高的POCD發(fā)生率,術(shù)中血糖波動幅度大和術(shù)后發(fā)生IR是其危險因素,應該密切監(jiān)測血糖和IR,保持術(shù)中血糖平穩(wěn)和減輕術(shù)后IR。但本研究中的病例均來自同一醫(yī)院,在病例的選擇上可能存在一定的偏倚,還需要擴大樣本范圍,進行多中心的研究,更深入探討術(shù)中血糖波動幅度和術(shù)后IR對老年全麻胸腔鏡肺癌根治術(shù)患者POCD的影響,為POCD的預防提供更多的參考依據(jù)。
參考文獻
[1] ACKENBOM M F,BUTTERS M A,DAVIS E M,et al. Incidence of postoperative cognitive dysfunction in older women undergoing pelvic organ prolapse surgery[J]. Int Urogynecol J,2021,32(2):433-442. doi:10.1007/s00192-020-04495-y.
[2] WANG W,MA Y,LIU Y,et al. Effects of dexmedetomidine anesthesia on early postoperative cognitive dysfunction in elderly patients[J]. ACS Chem Neurosci,2022,13(15):2309-2314. doi:10.1021/acschemneuro.2c00173.
[3] ENGIN M,GUVENC O. Many factors affect postoperative cognitive functions[J]. J Card Surg,2022,37(12):5702. doi:10.1111/jocs.17034.
[4] URCUN YS,ALTUN Y,PALA A A. Early and late predictors of postoperative neurocognitive dysfunction in cardiac surgery[J]. Ideggyogy Sz,2022,75(7-08):231-240. doi:10.18071/isz.75.0231.
[5] KURRA R JR,MADHUSUDHANA R. Impact of different crystalloids on the blood glucose levels of nondiabetic patients undergoing major elective surgeries[J]. Cureus,2023,15(1):e34294. doi:10.7759/cureus.34294.
[6] CHOI Y S,CHO B W,KIM H J,et al. Effect of preoperative oral carbohydrates on insulin resistance in older adults who underwent total hip or knee arthroplasty:a prospective randomized trial[J]. J Am Acad Orthop Surg,2022,30(20):971-978. doi:10.5435/JAAOS-D-21-00656.
[7] MENG Q Z,WANG Y,LI B,et al. Relationship between glycemic variability and cognitive function in lacune patients with type 2 diabetes[J]. World J Clin Cases,2023,11(5):1019-1030. doi:10.12998/wjcc.v11.i5.1019.
[8] CUI Y,TANG T Y,LU C Q,et al. Insulin resistance and cognitive impairment:evidence from neuroimaging[J]. J Magn Reson Imaging,2022,56(6):1621-1649. doi:10.1002/jmri.28358.
[9] 劉波. 胰島素抵抗與骨質(zhì)疏松椎體性壓縮骨折椎體成形術(shù)后新發(fā)骨折風險的相關(guān)性研究[D]. 瀘州:西南醫(yī)科大學,2018. LIU B. The correlation between insulin resistance and the risk of new fractures after vertebroplasty for osteoporotic vertebral compression fractures[D]. Luzhou:Southwest Medical University,2018.
[10] 唐東彪,田冶. Narcotrend監(jiān)測下麻醉深度對老年患者胃腸道腫瘤手術(shù)后認知功能障礙的影響[J]. 中國病理生理雜志,2020,36(3):475-478. TANG D B,TIAN Y. Effect of anesthesia depth with Narcotrend monitoring on cognitive function after gastrointestinal tumor surgery in elderly patients[J]. Chinese Journal of Pathophysiology,2020,36(3):475-478. doi:10.3969/j.issn.1000-4718.2020.03.0014.
[11] TJEERTES E,VAN FESSEM J,MATTACE-RASO F,et al. Influence of frailty on outcome in older patients undergoing non-cardiac surgery:a systematic review and Meta-analysis[J]. Aging Dis,2020,11(5):1276-1290. doi:10.14336/AD.2019.1024.
[12] 蘇容容,劉長紅. 胸腔鏡肺癌根治術(shù)后認知功能障礙對患者康復效果的影響[J]. 中國醫(yī)學前沿雜志(電子版),2020,12(11):53-57. SU R R,LIU C H. Effect of postoperative cognitive dysfunction on rehabilitation of patients undergoing thoracoscopic radical resection of lung cancer[J]. Chinese Journal of the Frontiers of Medical Science(Electronic Version),2020,12(11):53-57. doi:10.12037/YXQY.2020.11-09.
[13] 魏巍,鄭曦,谷宇,等. 術(shù)后鎮(zhèn)痛策略對老年單肺通氣患者術(shù)后神經(jīng)認知功能及早期康復的影響:90例隨機對照臨床試驗[J]. 南方醫(yī)科大學學報,2020,40(12):1821-1825. WEI W,ZHENG X,GU Y,et al. Effects of different postoperative analgesic strategies on postoperative neurocognitive function and quality of recovery in elderly patients undergoing one lung ventilation[J]. Journal of Southern Medical University,2020,40(12):1821-1825. doi:10.12122/j.issn.1673-4254.2020.12.19.
[14] YONG R,MENG Y. Preoperative neutrophil-lymphocyte ratio,an independent risk factor for postoperative cognitive dysfunction in elderly patients with gastric cancer[J]. Geriatr Gerontol Int,2020,20(10):927-931. doi:10.1111/ggi.14016.
[15] YOU H,HOU X,ZHANG H,et al. Effect of glycemic control and glucose fluctuation on in-hospital adverse outcomes after on-pump coronary artery bypass grafting in patients with diabetes: a retrospective study[J]. Diabetol Metab Syndr,2023,15(1):20. doi:10.1186/s13098-023-00984-4.
[16] 魯國,黃新穎,劉宏偉. 血糖變異性與急性腦梗死伴糖尿病患者短期認知功能的相關(guān)性分析[J]. 中國現(xiàn)代醫(yī)學雜志,2018,28(28):61-66. LU G,HUANG X Y,LIU H W. Association of blood glucose variability with short-term cognitive function in patients with acute cerebral infarction complicated with diabetes mellitus[J]. China Journal of Modern Medicine,2018,28(28):61-66. doi:10.3969/j.issn.1005-8982.2018.28.011.
[17] LIM J S,KIM C,OH M S,et al. Effects of glycemic variability and hyperglycemia in acute ischemic stroke on post-stroke cognitive impairments[J]. J Diabetes Complications,2018,32(7):682-687. doi:10.1016/j.jdiacomp.2018.02.006.
[18] MASCARENHAS FONSECA L,STRONG R W,SINGH S,et al. Glycemic variability and fluctuations in cognitive status in adults with type 1 diabetes (glucog):observational study using ecological momentary assessment of cognition[J]. JMIR Diabetes,2023,8:e39750. doi:10.2196/39750.
[19] 王敏,秦潔,李嵐昕,等. 2型糖尿病增加帕金森病風險相關(guān)機制研究進展[J]. 中國現(xiàn)代神經(jīng)疾病雜志,2023,23(6):566-570. WANG M,QIN J,LI L X,et al. Advance on mechanism of type 2 diabetes mellitus increasing the risk of Parkinson's disease[J]. Chin J Contemp Neurol Neurosurg,2023,23(6):566-570. doi:10.3969/j.issn.1672-6731.2023.06.016.
[20] ZHANG S,HE L,YU Y,et al. Effects of pre-operative oral carbohydrates on insulin resistance and postoperative recovery in diabetic patients undergoing coronary artery bypass grafting:study protocol for a prospective, single-blind, randomized controlled trial[J]. Trials,2022,23(1):1067. doi:10.1186/s13063-022-07042-w.
[21] MOUSA S,SAIF A,F(xiàn)ATHY M,et al. Assessment of early vascular changes in adult females with polycystic ovary syndrome: correlation with insulin resistance[J]. Gynecol Endocrinol,2023,39(1):2210226. doi:10.1080/09513590.2023.2210226.
[22] LIU J,KONG D,AI D,et al. Insulin resistance enhances binge ethanol-induced liver injury through promoting oxidative stress and up-regulation CYP2E1[J]. Life Sci,2022,303:120681. doi:10.1016/j.lfs.2022.120681.
[23] HAN X D,LI Y J,WANG P,et al. Insulin resistance-varying associations of adiposity indices with cerebral perfusion in older adults:a population-based study[J]. J Nutr Health Aging,2023,27(3):219-227. doi:10.1007/s12603-023-1894-2.
[24] XU J,GAO H,ZHANG L,et al. Melatonin alleviates cognition impairment by antagonizing brain insulin resistance in aged rats fed a high-fat diet[J]. J Pineal Res,2019,67(2):e12584. doi:10.1111/jpi.12584.