Meta-analysis of risk factors for new-onset diabetes mellitus after kidney transplantation

doi:10.3969/j.issn.2095-4344.1176

Abstract

Abstract:

BACKGROUND: Age, sex, body mass index, hepatitis C infection, immunosuppressive drugs and family history of diabetes mellitus are shown to be risk factors for new-onset diabetes mellitus after kidney transplantation, but their effects remain controversial.
OBJECTIVE: To systematically assess the risk factors for new-onset diabetes mellitus after kidney transplantation, so as to provide evidences for preventing and controlling the disease．
METHODS: PubMed, Embase, Cochrane Library and CBMdisc databases were searched for the articles concerning risk factors for new-onset diabetes mellitus after kidney transplantation published between January 2005 and May 2018. Two researchers extracted data from each study based on inclusion and exclusion criteria. Quality assessment was conducted in accordance with New castle-Ottawa Scale standard. Meta-analysis was performed on Revman 5.3 software to identify the risk factors for new-onset diabetes mellitus after kidney transplantation.
RESULTS AND CONCLUSION: (1) Twenty-one studies involving 8 206 patients were included. There were 1 489 cases of new-onset diabetes mellitus after kidney transplantation, and the morbidity was 18.15%. (2) The meta-analysis identified the following seven significant risk factors, non-modifiable risk factors: age ≥ 50 years, and donor type; modifiable risk factors: body mass index ≥ 25 kg/m2, acute rejection, tacrolimus usage, hepatitis C infection and polycystic kidney. (3) Uncertain risk factor was family history of diabetes. (4) To conclude, age, donor type, body mass index ≥ 25 kg/m2, acute rejection, tacrolimus usage, hepatitis C infection and polycystic kidney are risk factors for new-onset diabetes mellitus after kidney transplantation. But whether the family history of diabetes mellitus is the risk factor remains uncertain.

中国组织工程研究杂志出版内容重点：组织构建；骨细胞；软骨细胞；细胞培养；成纤维细胞；血管内皮细胞；骨质疏松；组织工程

Key words: Kidney Transplantation, Diabetes Mellitus, Risk Factors

CLC Number:

R446

Yang Jin, Zhang Meixia, Yan Pei, Cheng Qiao, Li Jianzhen . Meta-analysis of risk factors for new-onset diabetes mellitus after kidney transplantation[J]. Chinese Journal of Tissue Engineering Research, 2019, 23(15): 2450-2460.

Figures/Tables 7

2.2 纳入研究的一般情况从纳入的21个研究中，提取Meta分析中所需的基本信息：作者、文章发表的年份、研究的时间段、研究总例数、研究类型等，见表1。文章发表的年份由2005年至2017年，而研究的年份跨度由1984年2月至2014年12月。纳入研究的总病例数为8 206例，其中 1 489例发生肾移植后新发糖尿病，总发病率为18.15%。 2.3 纳入研究的文献质量评价纳入的21篇文献中，3篇为队列研究，18篇为病例对照研究。根据按纽尔卡斯-渥太华文献质量标准，3篇文献质量等级为A，18篇为B，见表2。2.4 Meta分析结果运用Revman5.3软件计算各危险因素合并OR值及其95%CI。Meta分析结果显示，肾移植后新发糖尿病不可干预的危险因素有2个：年龄≥50岁、供肾来源，肾移植后新发糖尿病可干预的危险因素有5个：体质量指数≥25 kg/m2、急性排斥反应、使用他克莫司、丙型肝炎病毒感染、多囊肾，而糖尿病家族史尚不能确定是否为肾移植后新发糖尿病的危险因素。 2.4.1 不可干预的危险因素 (1)年龄：纳入2项病例对照研究[9-10]，各研究间没有异质性(P=0.78，I2=0%)。应用固定效应模式分析显示：年龄≥50岁(OR=3.68，95%CI：2.02-6.71，P < 0.000 1)，见图2。说明年龄≥50岁会增加肾移植后新发糖尿病发生的危险因素。 (2)供肾来源：共纳入7篇研究，其中2篇为回顾性队列研究[11-12]，5篇为病例对照研究[13-16，27]。根据不同的研究设计分别进行分析，2篇回顾性队列研究合并结果异质性较小(P=0.27，I2=17%)，应用固定效应模型分析，结果显示尸体供肾(OR=2.72，95% CI：1.65-4.48，P < 0.000 01)，见图3；5篇病例对照研究合并结果异质性较低(P=0.16，I2=40%)，应用固定效应模式分析显示，说明尸体供肾较活体供肾，会使肾移植后新发糖尿病发生的风险增加1.74倍(OR=1.74，95% CI：1.27-2.38，P=0.000 5)，见图4。以上结果提示，尸体供肾是肾移植后新发糖尿病危险因素。 (3)糖尿病家族史：纳入15项研究，其中3篇为回顾性队列研究[11-12，17]，12篇为病例对照研究[10，13，15-16，18-25]。根据不同的研究设计分别进行分析，3篇回顾性队列研究合并结果异质性很高(P < 0.000 1，I2=91%)，应用随机效应模型分析，结果显示，糖尿病家族史不是肾移植后新发糖尿病的危险因素(OR=1.26，95% CI：0.33-4.78，P=0.73)，见图5；12篇病例对照研究间异质性较高(P < 0.000 01，I2=90%)，应用随机效应模式进行分析，结果显示，糖尿病家族史是肾移植后新发糖尿病的危险因素(OR=6.00，95% CI：2.54-14.16，P < 0.000 1)，见图6。为了明确12项病例对照研究异质性的来源，按照国籍进行亚组分析，中国籍的有7篇[15-16，18-22]，非中国籍的有5项研究[10，13，23-25]，结果显示，国籍不是异质性来源。以上结果提示，尚不能确定糖尿病家族史是肾移植后新发糖尿病的危险因素。 (4)性别：共纳入13篇研究，其中3篇为回顾性队列研究[11-12，17]，10篇为病例对照研究[13-14，16，18，20，23，25-28]。根据不同的研究设计分别进行分析，3篇回顾性队列研究合并结果不存在异质性(P=0.97，I2=0%)，应用固定效应模型分析，结果显示性别不是肾移植后新发糖尿病的危险因素(OR=1.05，95% CI：0.76-1.44，P=0.76)，见图7；10篇病例对照研究合并结果异质性较低(P=0.34，I2=11%)，应用固定效应模式分析，结果显示性别不是肾移植后新发糖尿病的危险因素(OR=0.96，95%CI：0.82-1.13，P=0.64)，见图8。以上结果提示，性别不是肾移植后新发糖尿病危险因素。"

(4)丙型肝炎病毒感染：共纳入11项研究，其中3篇为回顾性队列研究[11，12，17]，8篇为病例对照研究[9，13-15，21，24-25，27]。根据不同的研究设计分别进行分析，3篇回顾性队列研究合并结果异质性较低(P=0.35，I2=4%)，应用固定效应模型分析结果显示：丙型肝炎病毒感染(OR=2.24，95%CI：1.35-3.72，P=0.002)，见图12；8篇病例对照研究合并结果异质性较低(P=0.16，I2=0%)，应用固定效应模式分析显示，丙型肝炎病毒感染(OR=2.30，95%CI：1.73-3.05，P < 0.000 01)，见图13。以上结果提示丙型肝炎病毒感染是肾移植后新发糖尿病的危险因素。 (5)巨细胞病毒感染：共纳入5篇研究，其中2篇为回顾性队列研究[11-12]，3篇为病例对照研究[16，21，26]。根据不同的研究设计分别进行分析，2篇回顾性队列研究合并结果异质性较高(P=0.008，I2=86%)，应用随机效应模型分析，结果显示巨细胞病毒感染(OR=3.45，95%CI：0.73-16.34，P=0.12)，见图14；3篇病例对照研究合并结果不存在异质性(P=0.16，I2=0%)，应用固定效应模式分析显示，巨细胞病毒感染(OR=1.47，95%CI：0.83-2.59，P=0.19)，见图15。以上结果提示，巨细胞病毒感染不是肾移植后新发糖尿病危险因素。 (6)多囊肾：纳入3项病例对照研究[10，23-24]，各研究结果存在异质性(P=0.04，I2=46%)，应用随机效应模式分析显示：多囊肾(OR=3.87，95%CI：1.81-8.24，P=0.000 5)，见图16，说明了多囊肾是肾移植后新发糖尿病的危险因素。 (7)术前高血压：纳入5项病例对照研究[10，13，16，23，28]，各研究结果没有异质性(P=0.006，I2=0%)。应用固定效应模式分析显示：术前高血压(OR=1.03，95%CI：0.83-1.27，P=0.80)，图17，提示了术前高血压不是肾移植后新发糖尿病的危险因素。 (8)吸烟史：纳入3项病例对照研究[14，16，27]，各研究结果不存在异质性(P=0.72，I2=0%)。应用固定效应模式分析显示：吸烟史(OR=1.06，95%CI：0.52-2.71，P=0.87)，见图18，提示了吸烟史不是肾移植后新发糖尿病的危险因素。 2.5 发表偏倚的分析采用漏斗图评价发表偏倚。对性别、使用他克莫司、丙型肝炎病毒感染、急性排斥反应、糖尿病家族史5个纳入10研究的危险因素进行发表偏倚的分析，根据漏斗图判断，除糖尿病病家族史外，其他的危险因素没有发表偏倚的出现。术前高血压、多囊肾、吸烟史等因素纳入的研究过少，不宜做漏斗图，不排除发表偏倚的存在的可能。 2.6 敏感性分析依次将各研究因素中大样本或结果差异较大的研究剔除后再次进行Meta分析，发现剔除后的OR值、95%CI及P值与未剔除时的结果非常接近。因此，此次Meta分析的合并结果基本可靠、稳定。"

[1]	Zhang Yu, Tian Shaoqi, Zeng Guobo, Hu Chuan. Risk factors for myocardial infarction following primary total joint arthroplasty [J]. Chinese Journal of Tissue Engineering Research, 2021, 25(9): 1340-1345.
[2]	Wang Haiying, Lü Bing, Li Hui, Wang Shunyi. Posterior lumbar interbody fusion for degenerative lumbar spondylolisthesis: prediction of functional prognosis of patients based on spinopelvic parameters [J]. Chinese Journal of Tissue Engineering Research, 2021, 25(9): 1393-1397.
[3]	Zhang Chao, Lü Xin. Heterotopic ossification after acetabular fracture fixation: risk factors, prevention and treatment progress [J]. Chinese Journal of Tissue Engineering Research, 2021, 25(9): 1434-1439.
[4]	Chen Ziyang, Pu Rui, Deng Shuang, Yuan Lingyan. Regulatory effect of exosomes on exercise-mediated insulin resistance diseases [J]. Chinese Journal of Tissue Engineering Research, 2021, 25(25): 4089-4094.
[5]	Gao Shan, Huang Dongjing, Hong Haiman, Jia Jingqiao, Meng Fei. Comparison on the curative effect of human placenta-derived mesenchymal stem cells and induced islet-like cells in gestational diabetes mellitus rats [J]. Chinese Journal of Tissue Engineering Research, 2021, 25(25): 3981-3987.
[6]	Su Baotong, Wang Hanyu, Xu Yilang, Xie Yajuan, Cheng Zhian. Construction of a Nomogram prediction model for postoperative delirium after hip fracture in the elderly based on medical records from a hospital [J]. Chinese Journal of Tissue Engineering Research, 2021, 25(24): 3844-3849.
[7]	Liu Chang, Li Datong, Liu Yuan, Kong Lingbo, Guo Rui, Yang Lixue, Hao Dingjun, He Baorong. Poor efficacy after vertebral augmentation surgery of acute symptomatic thoracolumbar osteoporotic compression fracture: relationship with bone cement, bone mineral density, and adjacent fractures [J]. Chinese Journal of Tissue Engineering Research, 2021, 25(22): 3510-3516.
[8]	Tian Lin, Shi Xiaoqing, Duan Zhenglan, Wang Kuan, Zhang Li, Wang Peimin. Efficacy and safety of transverse tibial bone transport technique in the treatment of diabetic foot:a meta-analysis#br# [J]. Chinese Journal of Tissue Engineering Research, 2021, 25(20): 3275-3280.
[9]	Xia Wenshen, He Renjiao, Ai Jinwei, Wang Jun, Li Desheng, Pei Bin. Stem cell transplantation for diabetic patients with lower-extremity arterial disease: a meta-analysis [J]. Chinese Journal of Tissue Engineering Research, 2021, 25(19): 3110-3116.
[10]	Zhang Mingyan, Liu Xiangyang, Chang Lei, Chen Jing, Shen Xiongjie, Liu Bin, Peng Shuai, Zhang Chao, Wu Huanyu, Zhu Feng, Mou Haipin . Relationship between the intraoperative endplate injury and cage retropulsion after lumbar interbody fusion [J]. Chinese Journal of Tissue Engineering Research, 2021, 25(18): 2856-2862.
[11]	Luo Yicai, Li Hao. Effect of enhanced aryl hydrocarbon receptor expression on inflammatory response and healing of alveolar bone defects in diabetic rats [J]. Chinese Journal of Tissue Engineering Research, 2021, 25(14): 2166-2171.
[12]	Wang Xiaobo, Wang Changan, Han Jianle, Yang Qingyan, Yang Shuaiping, Yang Junwei. Influence of conversion from cyclosporine to tacrolimus on glucose metabolism and cardiovascular risk profiles in stable kidney transplant patients [J]. Chinese Journal of Tissue Engineering Research, 2021, 25(14): 2236-2240.
[13]	Zhang Tengfei, Wang Kun, Zhu Yanyu, Mei Wei, Wang Qingde. Meta-analysis of risk factors associated with adjacent segment degeneration after lumbar posterior fusion [J]. Chinese Journal of Tissue Engineering Research, 2021, 25(12): 1936-1943.
[14]	Liu Junchang, Gao Xiaolin, Jiang Taimao. Correlation of CY3A5 genetic polymorphism with concentration/dosage of tacrolimus and individualized administration of tacrolimus after kidney transplantation [J]. Chinese Journal of Tissue Engineering Research, 2021, 25(11): 1740-1744.
[15]	Guo Xuan, Xie Jun, Suo Jinrong, Li Yingrui, Huang Lei, Ma Munan, Li Jingjing, Fu Songtao. Transplantation of islet-like cells induced by human umbilical cord mesenchymal stem cells via different ways for the treatment of type 1 diabetic mice [J]. Chinese Journal of Tissue Engineering Research, 2021, 25(1): 78-83.