In this retrospective study, we reviewed all patients who received kidney transplants between January 2008 and December 2011. Patients who were diagnosed with DM before transplantation and received living donor kidney transplants were included in this study. Patients who were under the age of 18 years and those who received other organ transplants were excluded. ABO incompatible or human leukocyte antigen (HLA) sensitized kidney transplantation were also excluded. In addition, only patients who were followed up for at least 1 year after transplantation were included. Each patient from the study group was age- and sex-matched with two non-diabetic patients who received living donor kidney transplants during the same period. The outcomes of living donor kidney transplantation were compared between diabetic and non-diabetic patients. The study was approved by Asan Medical Center Institutional Review Board (2016-0208).

Patients received induction therapy with 20 mg of basiliximab on the day of surgery and on postoperative day 4. Maintenance immunosuppressive regimen consisted of a calcineurin inhibitor (cyclosporine or tacrolimus), antimetabolites (mycophenolate mofetil or azathioprine), and a corticosteroid. Postoperatively, all patients received oral trimethoprim/sulfamethoxazole (80/400 mg) for 6 months for pneumocystis pneumonia (PCP) prophylaxis. Cytomegalovirus (CMV) prophylactics were given routinely for CMV immunoglobulin G (IgG)-negative recipients matched with CMV IgG-positive donors. In other cases, preemptive therapy was administered based on serial monitoring for CMV antigenemia. All recipients underwent pre-transplant evaluation of cardiovascular risk by electrocardiogram, echocardiography and thallium-201 myocardial single photon emission computed tomography. When significant coronary artery disease was suspected, coronary angiogram was performed after discussion with cardiologists.

Baseline characteristics of patients, including age, sex, body mass index (BMI), past medical history, dialysis, and medications, were collected. Donor information and data on transplantation procedures was also reviewed. The occurrence of rejection, infection, cardiovascular events, and malignancy were also identified. Graft failure was defined as the recommencement of dialysis or direct second transplantation, and delayed graft function was defined as the use of dialysis in the first postoperative week [15]. Cardiovascular diseases included coronary artery disease, peripheral artery disease, and cerebral infarction. CMV infection was defined as ≥ 50 CMV (+) cells/200,000 white blood cells using a CMV antigenemia assay and was treated with ganciclovir. When plasma BK virus (BKV) DNA levels were > 10,000 copies/mL, BKV infection was diagnosed regardless of the presence of nephropathy. Protocol biopsies were not performed. Allograft biopsy was performed when rejection was clinically suspected. Renal biopsies were evaluated by light, electron, and immunofluorescence microscopy. C4d staining was performed on all specimens, and rejection was assessed with the Banff classification. The estimated glomerular filtration rate (eGFR) was calculated using the isotope-dilution mass spectrometry-traceable Modification of Diet in Renal Disease equation as follows: eGFR (mL/min/1.73 m²) = 175 × serum creatinine^–1.154 × age^–0.203 × (0.742 if patient is female) × (1.212 if patient is black).

Data were expressed as mean ± standard deviation or number (percentage). For categorical variables, the chi-square test and Fisher exact test were used. Continuous variables were compared using the Student t test or Mann-Whitney U test. Graft survival and mortality rates were evaluated by the Kaplan-Meier method with log-rank test. Cox proportional hazard analyses were used to assess variables associated with graft survival, using hazard ratios (HRs) and 95% confidence intervals (CIs). Variables with a p value of < 0.1 in the univariate analysis were included in the initial step of multivariate analysis. All reported p values were two-sided, and p values of < 0.05 were considered significant. All statistical analyses were performed using SPSS version 21.0 (IBM Corp., Armonk, NY, USA).

A total of 887 patients received kidney transplantation between 2008 and 2011 at Asan Medical Center (Fig. 1), of which 129 patients were excluded and 112 diabetic patients were included. However, four diabetic patients could not be matched with non-diabetic patients within 5 years of difference in age. In addition, 19 patients who received ABO incompatible or HLA sensitized kidney transplantation were excluded. Therefore, in the final analysis, 89 diabetic patients were compared with 178 non-diabetic patients. The mean BMI of diabetic patients was higher than that of non-diabetic patients (24.52 ± 3.68 kg/m² vs. 23.38 ± 2.89 kg/m², p = 0.006) (Table 1). Additionally, more diabetic patients than non-diabetic patients were diagnosed with cardiovascular disease before transplantation (21.3% vs. 6.7%, p = 0.001). The number of patients who were taking statin or renin-angiotensin blocker was not different between groups, but more patients were taking aspirin in diabetic patients (31.5% vs. 9.6%, p < 0.001). Other baseline characteristics were not significantly different between the two groups.

The incidence of acute rejection was not different between the diabetic and non-diabetic patients (Table 2). Delayed graft function occurred at similar rates in both groups (1.1% vs. 1.1%, p = 1). The overall incidence of infectious complications within the first year after transplantation did not differ significantly between the two groups (Table 3). However, urinary tract infections and other infections occurred more frequently after the first year of transplantation in diabetic patients than in non-diabetic patients (12.4% vs. 3.9%, p = 0.017; 14.6% vs. 2.8%, p < 0.001, respectively). Finally, diabetic and non-diabetic patients showed similar incidence rates of CMV and BK virus infection.

The incidence of graft failure was 6.7% in diabetic patients and 2.8% in non-diabetic patients. In addition, graft survival rates at 1, 3, and 5 years were 100%, 100%, and 95% in diabetic patients and 100%, 99%, and 98% in non-diabetic patients. The cause of graft failure was rejection in five of six graft failures in diabetic patients and four of five graft failures in non-diabetic patients. One patient in diabetic group experienced graft failure due to calcineurin inhibitor toxicity and rejection. The cause of graft failure was unknown in one patient in non-diabetic group. Both death-censored graft survival and non-death-censored graft survival were not significantly different between the two groups (p = 0.103 and p = 0.257 by log rank test) (Fig. 2). DM was not a significant risk factor for graft failure in univariate analysis (HR, 2.589; 95% CI, 0.789 to 8.490; p = 0.117) (Table 4). As summarized in Table 4, univariate analysis revealed that the risk of graft failure was significantly increased with HLA mismatches (4 to 6 mismatches) (HR, 4.794; 95% CI, 1.272 to 18.074; p = 0.021), acute rejection within the first year after transplantation (HR, 4.889; 95% CI, 1.280 to 18.666; p = 0.020) and acute rejection after the first year of transplantation (HR, 59.684; 95% CI, 7.638 to 466.342; p < 0.001). However, in multivariate analysis, a significantly higher risk of graft failure was observed with only acute rejection after the first year of transplantation (HR, 56.082; 95% CI, 7.169 to 438.702; p < 0.001) and high HLA mismatches (HR, 4.159; 95% CI, 1.100 to 15.722; p = 0.036). Mortality was not significantly different between diabetic and non-diabetic patients (0 vs. 2, p = 0.344 by log-rank test) (Fig. 3). One non-diabetic patient died due to progression of small cell lung cancer complicated with pneumonia 5 years and 9 months after transplantation, and the other died due to PCP complicated with candida sepsis 1 year and 2 months after transplantation.

Allograft function was not significantly different between diabetic patients and non-diabetic patients until 6 years after transplantation. More patients had proteinuria ≥ 2+ on dipstick in diabetic group after 5 years of transplantation, but no significant difference was shown in other periods. Allograft function and proteinuria in each group are summarized in Table 5.

More diabetic patients had cardiovascular disease before transplantation than non-diabetic patients. Likewise, diabetic patients experienced cardiovascular events more often than non-diabetic patients after transplantation (5.6% vs. 0.6%, p = 0.017). Three of the five diabetic patients and one non-diabetic patient who experienced cardiovascular events after transplantation already had past histories of cardiovascular disease before transplantation. Two of the diabetic patients had coronary artery disease and three had both coronary artery disease and peripheral artery disease. In contrast, a non-diabetic patient had only coronary artery disease. None of the patients in this study experienced stroke after transplantation. Finally, the incidence of malignancy of any type was not different between the two groups (4.5% vs. 2.8%, p = 0.487).

Among 178 non-diabetic patients, new-onset diabetes after transplantation (NODAT) developed in 62 patients (34.8%) during follow-up. Graft survival of diabetic patients and non-diabetic patients with or without NODAT was not significantly different (p = 0.142 and p = 0.217 by log rank test) (Fig. 4). In the subgroup analysis of diabetic patients and non-diabetic patients without NODAT, the incidence of rejection and infection showed similar pattern with the results of overall analysis. The incidences of acute rejection within 1 year and after 1 year of transplantation were not different between the two groups (p = 0.623 and p = 0.855). In addition, although total incidence of infection within 1 year did not show significant difference (32.6% vs. 26.7%, p = 0.439), sepsis occurred in more patients in diabetic patients compared with non-diabetic patients without NODAT (9.0% vs. 0.9%, p = 0.011). After 1 year of transplantation, the difference of overall incidence of infection between the two groups did not reach statistical significance (23.6% vs. 13.8%, p = 0.098). However, urinary tract infection and other infections occurred more often in diabetic patients after 1 year of transplantation (12.4% vs. 3.4%, p = 0.027 for urinary tract infection; 14.6% vs. 3.4%, p = 0.005). However, the incidence of CMV and BKV infection was not different between two groups.