Meta-analysis of risk factors for PTDM after renal transplantation in China in the past 10 years

XIA Man-cheng, ZHU Lian-yun, LV ding-yang, ZHOU Hui-yu, SHUANG Wei-bing?

1. Department of Urology, First Clinical Medical College, Shanxi Medical University, Taiyuan 030001, China

2. First Hospital of Shanxi Medical University, Taiyuan 030001, China

Keywords:Post-transplant diabetes mellitus Risk factors Protective factors Meta-analysis

ABSTRACT Objective: To systematically evaluate the main risk factors of post-transplant diabetes mellitus(PTDM) after renal transplantation in China in the past 10 years. Methods: CNKI, Wanfang,VIP, and PubMed were searched to collect the related literatures on risk factors of PTDM after renal transplantation published by Chinese scholars from January 2010 to October 2020. The data were extracted and Meta-analysis was performed by Revman 5.3 software.Results: A total of 18 case-control studies were included, involving 5 458 patients. There were 1 106 PTDM cases after kidney transplantation and 4 352 cases without PTDM after kidney transplantation. Meta-analysis results showed age [MD=6.09, P＜0.000 01], gender[OR=1.22, P=0.02], family history of diabetes [OR=5.56, P＜0.000 1], source of donor kidney[OR=1.87, P＜0.000 1], BMI [MD =1.76, P＜0.000 01], HBV infection [OR=2.52, P=0.04],HCV infection [OR=2.55, P＜0.000 1], CMV infection [OR=1.81, P=0.008], Cyclosporin A[OR=0.51, P=0.04], tacrolimus [OR=2.34, P=0.003], acute rejection [OR=2.72, P＜0.000 01], and smoking history [OR=2.01, P=0.000 6] were associated with PTDM after renal transplantation in China. Conclusion: Age, gender, family history of diabetes, source of kidney donors, BMI, HBV, HCV, CMV, tacrolimus, acute rejection, and smoking history are risk factors for PTDM after kidney transplantation in China. Cyclosporin A is protective factors of PTDM after kidney transplantation in China. These factors are worthy of attention by relevant clinical workers in our country.

1. Introduction

Post-transplant diabetes mellitus (PTDM) refers to patients who have not found abnormal blood glucose before successful organ transplantation, also known as NODAT. With the continuous development and improvement of renal transplantation in China,kidney transplantation has brought more gospel to more and more patients. But PTDM is often seen after renal transplantation. The incidence rate patients undergoing a kidney transplant is about 2%-50%[1-3], which can lead to inactivation of renal graft, cardiovascular events, infection and other related complications. It is harmful to the quality of life and prognosis of renal transplant patients. Studies have shown that the occurrence and development of PTDM after renal transplantation are affected by age, polycystic kidney disease,family history of diabetes, BMI, infection, immunosuppressive drugs, acute rejection and other risk factors [5-11], but there are many disputes among different studies. Due to the differences in race,gene and diet, it is necessary to conduct a meta-analysis on the risk factors of PTDM after renal transplantation in China, which will contribute to the targeted prevention and personalized management of PTDM after renal transplantation in China.

2. Data and methods

2.1 Inclusion and exclusion criteria

Inclusion criteria: (1) type and scope of study: case control studies published by Chinese scholars (including Hong Kong, Macao and Taiwan) at home and abroad in recent 10 years, Chinese scholars refer to the scholars of the first author / corresponding author unit in China (including Hong Kong, Macao and Taiwan); (2)Participants: patients aged 18 years or above who underwent renal transplantation for the first time; (3) Exposure factors: risk factors of PTDM after renal transplantation; (4) Outcome measures: PTDM occurred after renal transplantation; (5) The literature with definite diagnostic criteria for PTDM after renal transplantation should refer to the diagnostic criteria of who or American Diabetes Association.Exclusion criteria: (1) repetitive published literature; (2) Literature with too few risk factors or too small sample size; (3) The full text,incomplete data, data conversion and no control group were not available.

2.2 Retrieval strategy

CNKI, Wanfang, VIP, PubMed and other databases were searched to collect case-control studies on PTDM risk factors after renal transplantation published by Chinese scholars in domestic and foreign journals. The retrieval time was set from January 2010 to October 2020. It adopts the way of combining subject words and free words to search. Key words include: "kidney transplantation","diabetes", "new onset diabetes", "kidney transplantation", "renal transplantation", "diabetes mellitus", "PTDM", "NODAT", etc.

2.3 Literature screening and data extraction

Two researchers independently screened literature, extracted data and cross checked. If there are differences, they will be solved through discussion and consultation with the third researcher. When selecting the literature, the title should be read first. After excluding the obviously unrelated literature, the abstract and full text should be read further to determine whether to include. Data extraction included: (1) basic information of included studies: title, author, year of publication, study type, region of publication, follow-up time, etc;(2) The sample size of the study and the number of exposed and non exposed cases of the case group and the control group in each study;(3) The key elements of bias risk assessment; (4) Outcome measures and outcome measurement data.

2.4 Literature quality evaluation of included studies

Two researchers evaluated the quality of the included studies according to the New Castle Ottawa scale (NOS), and discussed and resolved the disagreement.

2.5 Statistical processing

Revman 5.3 software was used for meta analysis. Mean difference(MD) was used as the effect index for measurement data, and odds ratio (OR) was used as the effect index for count data. Point estimates and 95% confidence interval (CI) were calculated for each effect quantity. The heterogeneity between the results of the included studies was used χ2Inspection and analysis (the inspection level is α=1), and combined with I2to quantitatively determine the size of heterogeneity. If there was no statistical heterogeneity among the results, the fixed effect model was used for meta-analysis; On the contrary, the sources of heterogeneity were further analyzed.After excluding the influence of obvious clinical heterogeneity, the random effect model was used for meta-analysis. The test level of meta analysis was set as α= 0.05. Funnel plot was used to analyze publication bias.

3. Results

3.1 Literature screening process and results

A total of 590 relevant literatures were obtained by preliminary search. After layer by layer screening, 18 case-control studies [2,3,12-27] published by Chinese scholars at home and abroad in recent 10 years were included, including 5 458 patients, including 1 106 cases of PTDM after renal transplantation in China and 4 352 cases of no PTDM after renal transplantation in China. The process and results of literature screening are shown in Figure 1.

3.2 Basic characteristics of included studies and results of literature quality evaluation

Among the 18 included studies, there are 7 in northern China (north of Qinling Huaihe River) and 11 in southern China (south of Qinling Huaihe River); There were 7 articles with 6 scores, 10 articles with 7 scores and 1 article with 8 scores in NOS quality evaluation (Table 1).

3.3 Meta analysis results

According to the immutable risk factors and modifiable risk factors,the meta-study results of each factor were classified and summarized in Table 2 and Table 3.

3.3.1 Unalterable risk factors

(1) Age: 16 studies [3, 12-23, 25-27] were included, including 1 041 cases in PTDM group and 3 712 cases in non PTDM group. By heterogeneity test (P＜0.000 01; I2 =82% ) The results showed that age was a risk factor for PTDM after renal transplantation [MD=6.09, 95% CI (4.31, 7.88) years, P ＜ 0.000 01] (Figure 2).

Fig 1 Flowchart of literature screening

(2) Gender: 14 studies [2, 3, 12, 14-21, 23, 25, 27] were included,including 881 cases in PTDM group and 3 541 cases in non PTDM group. Heterogeneity test (P=0.12, I2 =32%) The fixed effect model showed that gender was a risk factor for PTDM after renal transplantation [OR=1.22, 95% CI (1.03, 1.43), P=0.02] (Figure 3).

(3) Family history of diabetes: 10 studies [3, 14-17, 20-23, 25, 26] were included, including 630 cases in PTDM group and 2 776 cases in non PTDM group. Heterogeneity test (P＜0.000 01, I2 =91%) The results showed that family history of diabetes was a risk factor for PTDM after renal transplantation [OR=5.56, 95% CI (2.54, 12.15),P＜0.000 1] (Figure 4).

(4) Polycystic kidney: 4 studies [3, 12, 14, 27] were included, including 238 cases in PTDM group and 988 cases in non PTDM group. The heterogeneity test (P=0.46, I2 =0%) The fixed effect model showed that polycystic kidney was not a risk factor for PTDM after renal transplantation [OR=1.45, 95% CI (0.81, 2.60), P=0.21] (Figure 5).

(5) Source of donor kidney: 6 studies [3, 12, 14, 17, 19, 21] were included, including 363 cases in PTDM group and 1 298 cases in non PTDM group. Heterogeneity test (P=0.14, I2=40%) The fixed effect model showed that the source of donor kidney was a risk factor for PTDM after kidney transplantation in China [OR=1.87,95% CI (1.37, 2.55), P＜0.000 1] (Figure 6).

3.3.2 Modifiable risk factors

(1) BMI: 16 studies [3, 12-23, 25-27] were included, including 1 041 cases in PTDM group and 3 712 cases in non PTDM group.Heterogeneity test (P＜0.000 01, I2 =87%) Using the random effect model, the results showed that BMI was a risk factor for PTDM after renal transplantation [MD=1.76, 95% CI (1.27, 2.25), P＜0.000 01](Figure 7).

Fig 2 Meta-analysis Forest plot of the association between age and PTDM after renal transplantation

Tab 1 Basic characteristics and NOS quality evaluation of the included literature

Tab 2 Summary of meta-analysis results of unmodifiable risk factors

(2) Hepatitis B virus (HBV) infection: 5 studies [3, 12, 16, 20, 21] were included, including 256 cases in PTDM group and 1 078 cases in non PTDM group. Heterogeneity test (P=0.02, I2=67%) The results showed that HBV infection was a risk factor for PTDM after renal transplantation [OR=2.52,95% CI (1.07, 5.95), P=0.04] (Figure 8).

(3) Hepatitis C virus (HCV) infection: 7 studies [12, 14, 16, 19-21, 27] were included, including 382 cases in PTDM group and 1 366 cases in non PTDM group. The heterogeneity test (P=0.32, I2=15%) The fixed effect model showed that HCV infection was a risk factor for PTDM after renal transplantation [OR=2.55, 95% CI (1.67, 3.88), P＜0.000 1] (Figure 9).

(4) Cytomegalovirus (CMV) infection: 7 studies [3, 12, 15, 16, 20, 21, 27]were included, including 428 cases in PTDM group and 1 537 cases in non PTDM group. The heterogeneity test (P=0.84, I2=0%) The fixed effect model showed that CMV infection was a risk factor for PTDM afterrenal transplantation [OR=1.81, 95% CI (1.28, 2.56), P=0.000 8](Figure 10).

Tab 3 Summary of meta-analysis results of modifiable risk factors

Fig 3 Forest plot of Meta-analysis on the association between gender and PTDM after renal transplantation

Fig 4 Forest map of Meta-analysis of the association between family history of diabetes and PTDM after renal transplantation

Fig 5 Forest plot of Meta-analysis of the association between polycystic kidney disease and PTDM after renal transplantation

Fig 6 Meta-analysis Forest plot of the association between donor kidney source and PTDM after renal transplantation

Fig 7 Forest plot of Meta-analysis of the association between BMI and PTDM after renal transplantation

Fig 8 Forest map of Meta-analysis of the association between HBV and PTDM after renal transplantation

Fig 9 Forest map of Meta-analysis of the association between HCV and PTDM after renal transplantation

Fig 10 Meta-analysis Forest map of the association between CMV and PTDM after renal transplantation

(5) Cyclosporine A: 8 case-control studies [2, 3, 14, 19, 20, 22, 24, 26]were included, including 406 cases in PTDM group and 1 869 cases in non PTDM group. Heterogeneity test (P＜0.000 01, I2=84%)Using the random effect model, cyclosporine A was a protective factor for PTDM after renal transplantation [OR=0.51, 95% CI (0.27,0.96), P=0.04] (Figure 11).

(6) Tacrolimus: 10 studies [3, 14, 15, 19, 20, 22, 24-27] were included,including 619 cases in PTDM group and 2 596 cases in non PTDM group. Heterogeneity test (P＜0.000 01, I2=87%) Using the random effect model, tacrolimus was a risk factor for PTDM after renal transplantation [OR=2.34, 95% CI (1.33, 4.12), P=0.003] (Figure 12).

(7) Acute rejection: 9 studies [3, 12, 15, 16, 18-21, 24] were included,including 414 cases in PTDM group and 1 758 cases in non PTDM group. Heterogeneity test (P=0.20, I2=28%) The fixed effect model showed that acute rejection was a risk factor for PTDM after renal transplantation [OR=2.72, 95% CI (2.05, 3.62), P＜0.000 01] (Figure 13).

(8) Hypertension: 3 studies [2,3,17] were included, including 205 cases in PTDM group and 794 cases in non PTDM group.Heterogeneity test (P=0.30, I2=17%) The fixed effect model showed that hypertension was not a risk factor for PTDM after renal transplantation [OR=1.16, 95% CI (0.81, 1.67), P=0.41] (Figure 14).

(9) Smoking history: 5 studies [3,12,19-21] were included, including 246 cases in PTDM group and 1 001 cases in non PTDM group.Heterogeneity test (P=0.17, I2=38%) The fixed effect model showed that smoking history was a risk factor for PTDM after renal transplantation [OR=2.01, 95% CI (1.35, 3.00), P=0.000 6] (Figure 15).

Fig 11 Forest plot of meta-analysis of the association between cyclosporin A and PTDM after renal transplantation

Fig 12 Meta-analysis Forest plot of the association between tacrolimus and PTDM after renal transplantation

Fig 13 Forest plot of Meta-analysis of the association between acute rejection and PTDM after renal transplantation

Fig 14 Forest plot of Meta-analysis of the association between hypertension and PTDM after renal transplantation

Fig15 Forest map of Meta-analysis on the association between smoking history and PTDM after renal transplantation

3.4 Evaluation of publication bias

Taking the factor of gender as an example, funnel plot was constructed, in which the scattered points were roughly symmetrical,indicating that the possibility of publication bias was low (Figure 16).

Fig 16 Funnel plot results for gender factors

4. Disscusion

The occurrence and development of PTDM after renal transplantation are affected by many factors, including immutable factors and modifiable factors. The immutable factors include age,family history of diabetes, polycystic kidney disease, race, gene polymorphism, etc. [28-34], and the modifiable factors include BMI,acute rejection, immunosuppressive agents, infection, etc. [8,10, 11,35].This study included age, gender, family history of diabetes mellitus,polycystic kidney, source of donor kidney, BMI, tacrolimus,cyclosporine A, acute rejection, HBV, HCV, CMV, hypertension,smoking history and other factors that may affect PTDM after renal transplantation in China.

Many studies have shown that age is an important risk factor for PTDM after renal transplantation [14,17,36]. This study also shows that age is a risk factor for PTDM after renal transplantation in China,which may be related to the increase of islet age β The function of cells decreased gradually [37].

At present, most studies have shown that gender is not a risk factor for PTDM after renal transplantation [2,12,21], but some scholars have shown that the risk of PTDM in male patients after renal transplantation is 1.81 times higher than that in female patients [23].After synthesizing 14 studies of Chinese scholars, this study found that gender is a risk factor for PTDM after renal transplantation in China, The risk of PTDM in male patients after renal transplantation is 1.22 times higher than that in female patients. More research is needed to explore the specific causes and pathogenesis.

Family history of diabetes is a risk factor that scholars at home and abroad have been concerned about [36]. It is reported that the risk of PTDM in Chinese patients with family history of type 2 diabetes after renal transplantation is higher than that in patients without family history of type 2 diabetes [25,27]. This study suggests that family history of diabetes can increase the risk of PTDM to 5.79 times, which is consistent with the previous research results.

Whether polycystic kidney is a risk factor for PTDM after renal transplantation remains controversial at home and abroad. Prakash et al. [30] showed that polycystic kidney is a risk factor for PTDM after renal transplantation, while Yang Jin et al. [3] showed that polycystic kidney is not a risk factor for PTDM after renal transplantation.Based on the combined analysis of four case-control studies published by Chinese scholars, this study shows that polycystic kidney is not a risk factor for PTDM after kidney transplantation in China, which may be related to the lack of research in this area by Chinese scholars. Therefore, more Chinese scholars are needed to improve and supplement the research in this area.

At present, a number of domestic studies have shown that the source of donor kidney is not a risk factor for PTDM after renal transplantation [3, 12, 19]. Compared with living donor kidney,cadaveric donor kidney does not increase the risk of PTDM after renal transplantation, but Chen MINLING and other studies in China have shown that cadaveric donor kidney can increase the risk of PTDM after renal transplantation to 3.65 times [21]. This metaanalysis showed that the source of donor kidney is a risk factor for PTDM after kidney transplantation in China, which may be related to the high dosage of hormone in cadaveric donors.

Many studies have shown that BMI is a risk factor for PTDM after renal transplantation [9,15,29], and its mechanism may be related to obesity stimulating islets β It is related to insulin resistance of cells,which leads to the decrease of blood glucose clearance ability [7, 38].The results of 16 case-control analysis showed that BMI was a risk factor for PTDM after renal transplantation in China, which was consistent with the previous literature reports.

Infection has always been a kind of risk factors related to PTDM after renal transplantation, including HBV, HCV and CMV infection.The pathogenesis may be related to the effect of infection on islets β It is related to direct cell injury, decreased glucose uptake and glycogen production in liver, and increased insulin resistance caused by insulin receptor deficiency [39]. This study shows that HBV, HCV and CMV are risk factors for PTDM after renal transplantation in China.

Immunosuppressive agents are a kind of risk factors for PTDM.Currently, the immunosuppressive agents reported to cause PTDM include calcineurin inhibitors, corticosteroids, rapamycin target protein inhibitors, etc. [8,11,27]. There are many studies on calcineurin inhibitors such as tacrolimus and cyclosporine A in China. Therefore,this study conducted a meta-analysis on tacrolimus and cyclosporine A. the combined results showed that tacrolimus was a risk factor for PTDM after renal transplantation in China, while cyclosporine A was a protective factor for PTDM after renal transplantation in China, This suggests that cyclosporin A should be used as much as possible in the selection of immunosuppressants.

Acute rejection is currently recognized as a risk factor for PTDM in renal transplantation [3,11,40]. On the one hand, it may be related to the high-dose use of glucocorticoid after acute rejection. At this time, the patient will develop insulin resistance due to the effect of glucocorticoid [41,42]. On the other hand, acute rejection can make glucocorticoid, growth hormone, glucagon The level of catecholamine and other insulin antagonists is increased, which leads to the increase of blood glucose level [19]. The combined analysis of nine case-control studies in China shows that acute rejection can increase the risk of PTDM after kidney transplantation by 2.72 times.

At present, there are few reports about the effect of hypertension on PTDM after renal transplantation in China. From the theoretical mechanism, the elevation of homocysteine in patients with hypertension can cause the secretion of a variety of inflammatory factors, which can make the function of adipose tissue abnormal,promote the production and secretion of resistin, and then lead to inflammatory reaction and insulin resistance [43-45]. Our study included three domestic literatures, and the combined results showed that hypertension was not a risk factor for PTDM after renal transplantation in China, so more studies are needed to explore. In addition, this study also found that smoking history is a risk factor for PTDM after kidney transplantation in China, which suggests that patients undergoing kidney transplantation in China should quit smoking.

Research limitations: the type of literature included in this study is case-control study, and the demonstration intensity is lower than that of randomized controlled study, cohort study and other research types. The sample size of some literatures included in this study is small or the number of risk factors analyzed is small, which causes some uncertainty to explore the relationship between PTDM and renal transplantation in China. Therefore, in the future, more institutions in China need to carry out large sample and more factors related research to further clarify the related risk factors of PTDM after kidney transplantation in China.

In conclusion, age, gender, family history of diabetes, source of donor kidney, BMI, HBV, HCV, CMV, tacrolimus, acute rejection and smoking are risk factors for PTDM after renal transplantation in China. Cyclosporine A is a protective factor for PTDM after renal transplantation in China. Clinical medical workers should pay attention to these factors. In addition, Chinese scholars need to carry out more relevant studies to further confirm the influence of these factors.

Author's contribution:

Xia Man-cheng is responsible for the design, implementation, data processing, writing and revision of the research; Zhu Lian-yun, Lv Ding-yang and Zhou Hui-yu are responsible for data collection and article revision; The double guards are responsible for reviewing the content of the article and providing administrative support.

All authors declare that there is on conflict of interest.