Post-transplant diabetes mellitus in kidney transplant recipients: A prospective cohort study of risk factors and clinical outcomes

Abstract

Post-transplant diabetes mellitus (PTDM) is a common metabolic complication following kidney transplantation and is associated with an increased risk of graft failure, cardiovascular disease, infections, and mortality. We prospectively analyzed the risk factors and clinical consequences of PTDM in patients who underwent kidney transplantation at our center. The study included 130 transplant recipients without preexisting diabetes (79 males, 51 females; mean age: 47.6 years). PTDM was defined as the requirement for pharmacologic treatment at least 3 months after transplantation or based on the oral glucose tolerance test. The observation period ranged from 4 to 75 months. The incidence of PTDM was 21.5% (28/130). Patients who developed PTDM were older (P = .02) and had a higher body mass index (P = .04). Median fasting glucose and HbA1c levels prior to transplantation were significantly higher in patients who later developed PTDM. No significant differences were observed in sex, Charlson Comorbidity Index, dialysis method or duration, number of HLA mismatches, frequency of acute rejection, delayed graft function, reoperation, graft function, steroid dose, or baseline serum levels of insulin, C-peptide, lipid profile, or markers of insulin resistance. Tacrolimus was part of the immunosuppressive regimen in 89% of PTDM patients and 78% of non-PTDM patients (P > .05). In this cohort, general risk factors for diabetes appeared to play a greater role in the development of PTDM than transplantation-specific variables. Simple pre-transplant indicators such as blood glucose and HbA1c may help identify patients at increased risk of PTDM.

1. Introduction

Post-transplant diabetes mellitus (PTDM), defined as newly diagnosed diabetes mellitus in the post-transplant setting,^[1] is associated with an increased risk of graft failure^[2,3] and, in several studies, with cardiovascular disease and mortality.^[2,4–7] Patients with PTDM may develop microangiopathies at an accelerated rate compared to those with non-transplant-related diabetes^[8] and are at increased risk of cardiac events, with incidence rates comparable to patients with diabetes prior to transplantation.^[5] Early detection and treatment of PTDM can help mitigate its long-term consequences.

The International Consensus Guidelines on New-Onset Diabetes After Transplantation (NODAT), published in 2003, recommended using the American Diabetes Association criteria for the diagnosis of type 2 diabetes.^[9,10] Updated recommendations from an international consensus meeting in 2014^[1] proposed reverting the terminology from “New-Onset Diabetes After Transplantation” to “Post-transplant Diabetes Mellitus (PTDM)” to encompass patients with previously undiagnosed diabetes.

Although the oral glucose tolerance test (OGTT) is somewhat impractical in routine clinical settings, it remains the gold standard for diagnosing PTDM. Impaired fasting glucose (IFG) and impaired glucose tolerance (IGT) are considered prediabetic conditions and are associated with an increased risk of developing diabetes. Performing an OGTT prior to kidney transplantation (KTx) is useful and recommended.^[11] Pre-transplant IGT has been strongly associated with PTDM, increasing the risk by approximately 3.8 times.^[12]

Recent studies have reported that PTDM occurs in 8% to 39% of patients within the first year following KTx.^[13–17] OGTT should be incorporated into clinical practice once patients are stable on maintenance immunosuppression, exhibit stable graft function, and are free from acute infections.^[1,9,10,18]

Risk factors for PTDM include both general diabetes risk factors and transplant-specific factors, particularly immunosuppressive therapy. Understanding these risk factors is essential for prevention, risk stratification, and optimizing immunosuppressive protocols. Immunosuppression should primarily be tailored to the patient’s immunological risk to prevent graft rejection.^[10,13,19] Appropriate hypoglycemic therapy should be initiated in all cases of hyperglycemia, regardless of the immunosuppressive regimen. Notably, early insulin therapy has been shown to reduce the risk of PTDM by decreasing β-cell stress and glucotoxicity.^[20]

The aim of this study was to evaluate clinical and biochemical risk factors for PTDM present prior to KTx and to assess early post-transplant clinical consequences of PTDM among patients treated at the Gdańsk Transplantation Center.

2. Materials and methods

We prospectively analyzed the risk factors and clinical consequences of PTDM in a representative cohort of Caucasian patients who underwent KTx at our center between 2016 and 2022. Between November, 2016 and December, 2022, a total of 658 patients underwent KTx, including 600 individuals without a prior diagnosis of diabetes mellitus. The study included 130 transplant recipients (79 males and 51 females; mean age: 47.6 years) without preexisting diabetes. All eligible and consecutively transplanted patients meeting the inclusion criteria within the study period were analyzed. Patients for whom the required laboratory tests could not be performed due to timing or logistical issues, those receiving ongoing systemic corticosteroid therapy prior to transplantation, and those who refused to participate were excluded from the study. Figure 1 presents a flowchart of the study. All but 1 patient received a cadaveric graft. The observation period ranged from 4 to 75 months. Baseline clinical and transplant-related characteristics of the study cohort are summarized in Table 1.

Figure 1.

Flowchart of the study cohort.

Table 1.

Characteristics of patients with and without PTDM.

Variable	No PTDM group n = 102	PTDM group n = 28	Total n = 130	P value
Age (yr) range min-max; median (IQR)	18–74.5; 45.85 (36.225, 55.3)	25-71; 55.35 (41.75, 64.62)	18–74.5 47.81 (37–58)	.019
BMI (kg/m2) median (IQR)	24.35 (22.3, 27.812)	26.8 (24.225, 29.325)	24.7 (22.58–28.7)	.044
BMI > 30 kg/m2, n (%)	11 (10.8)	6 (21.4)	17 (13.1)	.12
Gender – male n (%)	63 (61.8)	16 (57.1)	79 (60.8)	.408
2nd and 3rd KTx n (%)	9 (8.8)	1 (3.6)	10 (7.7)	.32
Dialysis modality before KTx n (%) HD PD PREE	71 (69.6) 21 (20.6) 10 (9.8)	20 (71.4) 6 (21.4) 2 (7.2)	91 (70) 27 (20.8) 12 (9.2)	.526 .55 .5
Pre-transplant dialysis time (mo) median (IQR)	20.295 (11.242, 46.59)	24.5 (11.75, 35.795)	20.98 (11.0, 40.549)	.318
MM median (IQR)	3.00 (2.00, 3.00)	3.00 (2.00, 3.00)	3 (2.00, 3.00)	.656
Cause of ESRD n (%) GN ADPKD HN IN Not known Other	55 (53.9) 12 (11.8) 8 (7.8) 12 (11.8) 10 (9.8) 5 (4.9)	11 (39.3) 7 (25) 3 (10.71) 2 (7.14) 2 (7.14) 3 (10.71)	66 (50.8) 19 (14.6) 11 (8.5) 14 (10.8) 12 (9.2) 8 (6.1)	.12 .08 .44 .38 .5 .23
IND use – Thymoglobuline®/Grafalon® or basiliximab-n-Simulect® (%)	34 + 32 = 66 (64.7)	6 + 16 = 22 (78.6)	40 + 48 = 88 (67.7)	.12
CSA use n (%)	22 (21.6)	3 (10.7)	25 (19.2)	.207
CSA blood level µg/L median (IQR)	234.5 (201.85–258.75)	222 (203–354.2)	223 (199.8–260)	.229
TAC use n (%)	80 (78.4)	25 (89.3)	105 (80.8)	.207
Tac blood level µg/L median (IQR)	12.65 (10.65–15.025)	11.7 (9.6–13.7)	12.6 (10.3–14.8)	.332
2-wk cumulative methylprednisolon dose (mg/kg)	3.021 (2.667, 3.556)	2.783 (2.590, 3.584)	2.977 (2.606, 3.556)	.327
AR n (%)	4 (3.9)	2 (7.1)	6 (4.6)	.478
DGF n (%)	23 (22.6)	4 (14.3)	27 (20.8)	.344
Duration of hospitalization after KTx days median (IQR)	15 (13–21)	15.5 (13–21)	15 (13–21)	.78
SCC (mg/dL) 1 mo after KTx median (IQR)	1.605 (1.145–1.985)	1.555 (1.158–1.933)	1.565 (1.14–1.94)	.750
eGFR 1 mo after KTx mL/min median (IQR)	57.797 (48.347–81.284)	59.599 (46.15–77.208)	58.54 (48.14–81.17)	.855
REOP n (%)	14 (13.7)	4 (14.3)	18 (13.8)	.576
CMV n (%)	9 (8.8)	2 (7.1)	11 (8.5)	.778
UTI n (%)	15 (14.7)	7 (25)	22 (16.9)	.203
Charlson Index median (IQR)	3 (2–3)	3 (2–4)	3 (2–4)	.656
WIT min median (IQR)	30 (26–37)	30 (27.5–40)	30 (26–37)	.179
CIT min median (IQR)	867 (668–1169)	795 (553–916)	829 (644.5–1169)	.073
Deaths n (%)	4 (3.9%)	2 (7.1%)	6 (4.6%)	.38
Graft loss n (%)	2 (2.0)	2 (7.1%)	4 (3.1%)	.2

ADPKD = autosomal dominant polycystic kidney disease, AR = acute rejection, BMI = body mass index, CIT = cold ischemia time, CMV = cytomegalovirus infection, CsA = cyclosporin A, DGF = delayed graft function, eGFR 4p MDRD = estimated glomerular filtration rate 4 point Modification of Diet in Renal Disease formula, F = female, GN = glomerulonephritis, HD = hemodialysis, HN = hypertensive nephropathy, IN = interstitial nephritis, IND = induction, IQR = inter-quartile range, KTx = kidney transplantation, M = male, MM = number of mismatches, PD = peritoneal dialysis, PREE = preemptive transplantation, REOP = reoperation, SCC = serum creatinine concentration, TAC = tacrolimus, UTI = urinary tract infection, WIT = warm ischemia time.

2.1. Inclusion and exclusion criteria

Inclusion criteria:

• Adult patients (≥18 years old) who underwent KTx at the Gdańsk Transplantation Center between November 2016 and December 2022.

• No documented diagnosis of diabetes mellitus prior to transplantation.

• Minimum post-transplant follow-up of 4 months.

• Signed informed consent and participation in the prospective observational study.

Exclusion criteria:

• Known diabetes mellitus prior to KTx.

• Inability to perform laboratory tests required for the study due to timing/logistics*.

• Chronic or ongoing systemic corticosteroid therapy prior to transplantation.

• Refusal to participate in the study.

• Graft loss within the first 3 months post-transplant.

*The vast majority of KTxs at our center are performed using organs from deceased donors. In accordance with national regulations in Poland, transplant recipients are transported to the donor’s transplant center to undergo the procedure, often on very short notice. As a result, patients may arrive at the transplant center at any time (day or night) depending on organ availability and logistical coordination. Due to the unpredictable nature of these admissions, it was not always feasible to ensure the presence of study personnel responsible for conducting additional pre-transplant blood testing. Although the study-related laboratory tests did not increase the overall volume of blood collected, they required specific handling and timely processing (e.g., for insulin level assessment), which could not be performed outside standard working hours or when study personnel were unavailable. Therefore, patients for whom proper pre-transplant blood processing could not be guaranteed were excluded from the study.

2.2. Diagnosis of PTDM

PTDM was defined as the requirement for pharmacologic therapy at least 3 months after transplantation or based on the results of an OGTT.

A standard 75 g OGTT was performed after a minimum of 8 hours of fasting. Blood samples were collected at 0 and 120 minutes to measure plasma glucose levels. The 2005 American Diabetes Association (ADA) criteria were used to define PTDM, IFG, and IGT. PTDM was defined as fasting plasma glucose > 126 mg/dL and/or 2-hour plasma glucose > 200 mg/dL. IFG was defined as fasting plasma glucose between 100 and 126 mg/dL, and IGT as 2-hour plasma glucose between 140 and 200 mg/dL.

2.3. Variables of interest

The following data were assessed as potential risk factors for PTDM: age, sex, cause of end-stage kidney disease, method and duration of pre-transplant dialysis, Charlson Comorbidity Index (CCI), body mass index (BMI), number of HLA mismatches, ischemic time, immunosuppressive medications, acute rejection (AR), delayed kidney graft function, reoperation, and laboratory tests: serum creatinine concentration, estimated glomerular filtration rate (eGFR), lipid profile, HbA1c, glucose, insulin, homocysteine, uric acid, C-peptide, albumin and insulin resistance indicators: HOMA-IR*, QUICKI**, TyG index***.

*HOMA-IR (HOmeostatic Model Assessment-Insulin Resistance) = fasting glucose (mg/dL) X fasting insulin (mU/L)/405.

**QUICKI Index (Quantitative Insulin Sensitivity Check Index) = 1/[log (fasting insulin, U/mL) + log (fasting glucose, mg/dL)].

*** TyG index (triglycerides and glucose index) = ln [(triglycerides (mg/dL) × glucose (mg/dL)/2)].

The following variables were assessed as potential risk factors for PTDM:

• Demographic and clinical variables:

Age, gender, BMI, cause of end-stage kidney disease, dialysis modality and duration, CCI, number of HLA mismatches, ischemic time, and immunosuppressive regimen were obtained from the institutional transplant database and medical records.

• Transplant-related outcomes:

AR episodes, delayed graft function (DGF), and reoperations were documented during hospitalization and follow-up visits and verified in electronic health records.

• Laboratory measurements:

Blood samples were collected pre-transplant and analyzed in the hospital’s central laboratory using standardized automated analyzers. All laboratory values were measured using the same equipment and protocols for all patients, regardless of eventual PTDM status.

2.4. Definition of comorbidities and clinical conditions

Comorbidities and transplant-related complications were identified based on a comprehensive review of medical records, transplant documentation, and laboratory/imaging results.

• CCI: Calculated using the original scoring system,^[21] which assigns weighted points to comorbid conditions (e.g., cardiovascular disease, chronic pulmonary disease, malignancy, liver disease) based on the patient’s documented medical history at the time of KTx.

• Urinary tract infection (UTI): Diagnosed based on clinical symptoms (e.g., dysuria, fever, urinary urgency) with laboratory confirmation of leukocyturia and/or positive urine culture (>10⁵ CFU/mL of a single uropathogen).

• Cytomegalovirus (CMV) infection: Defined by the presence of CMV pp65 antigenemia and/or detection of CMV DNA in blood via quantitative PCR, with or without associated clinical symptoms (e.g., fever, malaise, leukopenia, elevated liver enzymes). Testing was routinely performed as part of post-transplant viral monitoring.

• AR

Defined as a clinically suspected episode of immunologic graft injury, characterized by a ≥ 25% increase in serum creatinine from baseline in the absence of other identifiable causes (e.g., dehydration, infection, urinary tract obstruction, or nephrotoxic levels of calcineurin inhibitors). Due to the observational nature of this study and institutional practice patterns, biopsy confirmation was not consistently performed. The diagnosis was based on clinical signs (e.g., rising creatinine, graft tenderness, reduced urine output), an elevated resistance index on Doppler ultrasound, and the patient’s response to anti-rejection treatment (high-dose corticosteroids). All episodes of AR were reviewed retrospectively using medication records.

• DGF: Defined as the requirement for dialysis during the first 7 days following transplantation.

• Reoperation: Defined as any unplanned surgical procedure performed within 30 days after KTx due to transplant-related complications, such as vascular thrombosis, bleeding, lymphocele, wound dehiscence, or urinary tract leakage.

• Graft loss: Defined as return to chronic dialysis, graft nephrectomy.

2.5. Statistical analyses

The normality of quantitative data distributions was assessed using the Shapiro–Wilk W test. As many of the analyzed parameters exhibited non-Gaussian distributions, all quantitative data are presented as medians with corresponding inter-quartile ranges. Categorical variables are summarized as counts and percentages. The number of missing values for each variable is reported.

Between-group differences were evaluated using the Mann–Whitney U test for quantitative data and Fisher exact test for categorical variables. Univariate binary logistic regression was performed to assess the association between potential predictors and the development of diabetes following KTx. Results are presented as odds ratios (ORs) with 95% confidence intervals (CIs) and P-values. The P-values were calculated using the Wald test to evaluate the statistical significance of predictors.

To identify a multivariable logistic regression model, forward variable selection based on the Bayesian Information Criterion was applied. Additionally, elastic net regularization was used as an alternative model selection approach. Ultimately, only the univariate model with age as a predictor was retained as the final model.

All statistical analyses were performed using R software, version 4.3.3.

3. Results

3.1. Glycemic disturbances in patients after KTx

3.1.1. Incidence of PTDM and prediabetic states (IFG, IGT)

The incidence of PTDM was 21.5% (28/130), while PTDM requiring pharmacologic treatment was observed in 19.2% (25/130). Prediabetic states, including IFG and IGT, were diagnosed in 30 patients: IFG in 18 patients, IGT in 5 patients, and both IFG and IGT in 7 patients.

Overall, the prevalence of glycemic disturbances (PTDM + IFG + IGT) was 44.6% (58/130).

The mean observation period after KTx was 18 months (range: 4–75 months). The diagnosis of diabetes was made, on average, 4 months post-KTx (range: 1–19 months).

3.1.2. Antidiabetic treatment

Among the 28 patients receiving antidiabetic therapy, 23 were treated with intensive insulin therapy, 1 with metformin, and 4 with diet alone (without pharmacologic treatment).

3.2. Characteristics of KTx recipients in the PTDM and non-PTDM groups

3.2.1. Age

The median age of patients who developed PTDM was significantly higher (P = .02). The median age in the PTDM group was 55.35 years, compared to 45.85 years in the non-PTDM group (Table 1).

3.2.2. BMI

BMI was higher in patients who developed PTDM. The median BMI in the PTDM group was 26.8, compared to 24.35 in the non-PTDM group (P = .04; Table 1).

3.2.3. Non-significant factors

No statistically significant differences were found between the PTDM and non-PTDM groups with respect to sex, cause of ESRD, method and duration of pre-transplant dialysis, CCI, number of HLA mismatches, ischemia time, or the number of second and third kidney transplants (Table 1).

3.3. Immunosuppression

There were no significant differences in immunosuppressive protocols between patients with and without PTDM.

3.3.1. Calcineurin inhibitors

Tacrolimus was part of the immunosuppressive regimen in 25 of 28 PTDM patients (89%) and in 80 of 102 non-PTDM patients (78%); this difference was not statistically significant (P = .2). Tacrolimus trough levels were also not significantly different between the PTDM and non-PTDM groups (median: 11.7 vs 12.6 ng/mL; P = .3). Similarly, there was no significant difference in cyclosporine trough levels (median: 222 vs 234 µg/L; P = .2; Table 1).

3.3.2. Induction therapy

Induction therapy (Thymoglobulin®, Grafalon®, or Simulect®) was used in 88 of 130 patients (67.7%), including 22 of 28 PTDM patients (78.6%) and 66 of 102 non-PTDM patients (64.7%) (P = .12; Table 1).

3.3.3. Steroids

The 2-week cumulative methylprednisolone dose per kilogram of body weight did not differ significantly between the PTDM and non-PTDM groups (P = .3; Table 1).

3.4. Complications after KTx in patients with and without PTDM

3.4.1. AR

AR (not biopsy-proven) occurred in 6 of 130 patients (4.6%). The incidence was 2 of 28 (7.1%) in the PTDM group and 4 of 102 (3.9%) in the non-PTDM group (P = .48; Table 1).

3.4.2. DGF

Delayed graft function, defined as the need for dialysis within 1 week post-transplant, occurred in 20.8% of all patients: 14.3% in the PTDM group and 22.5% in the non-PTDM group (P = .34; Table 1). No statistically significant difference was observed.

3.4.3. UTI

UTI was diagnosed in 22 of 130 patients (16.9%). Among PTDM patients, the incidence was 25% (7 of 28), compared to 14.7% (15 of 102) in the non-PTDM group. The difference was not statistically significant (Table 1).

3.4.4. CMV infection

CMV infection occurred in 11 of 130 patients (8.5%), with no statistically significant difference between PTDM and non-PTDM groups (Table 1).

3.4.5. Reoperation

Reoperation due to surgical complications was necessary in 18 of 130 patients (13.8%). There was no significant difference between groups (Table 1).

3.4.6. Death and graft loss

There were 2 deaths in the PTDM group (caused by COVID-19 and sepsis) and 4 in the non-PTDM group (two due to COVID-19, 1 due to ruptured thoracic aortic aneurysm, 1 due to urosepsis) during the first year after KTx; the difference was not statistically significant.

Graft loss occurred in 2 non-PTDM patients (due to noncompliance and unknown cause) and in 2 PTDM patients (due to kidney graft vein thrombosis and urosepsis), with no statistically significant difference.

3.5. Kidney graft function

Serum creatinine concentration and eGFR (calculated using the Modification of Diet in Renal Disease formula) 1 month after KTx did not differ significantly between PTDM and non-PTDM patients. In the PTDM group, the median creatinine concentration was 1.555 mg/dL, and the median eGFR was 59.6 mL/min. In the non-PTDM group, the median creatinine concentration was 1.605 mg/dL, and the median eGFR was 57.8 mL/min (Table 1).

3.6. Laboratory tests performed directly before KTx: comparison between PTDM and non-PTDM groups

3.6.1. Statistically significant parameters: HbA1c and fasting glucose before KTx

3.6.1.1. HbA1c

The median HbA1c, expressed in both % and mmol/mol, was significantly higher in the PTDM group (Table 2). HbA1c values above the normal range (≥5.7%/ ≥39 mmol/mol) were significantly more common in patients who later developed PTDM (P = .007). However, none of the patients had HbA1c values ≥ 6.5% or ≥ 48 mmol/mol, which are diagnostic thresholds for diabetes (Table 2).

Table 2.

Laboratory tests before KTx in patients with and without PTDM.

Variable	No PTDM group n = 102	PTDM group n = 28	Total n = 130	P value
HbA1c %, median (IQR)	5.300 (5, 5.500)	5.55 (5.175, 5.8)	5.3 (5, 5.6)	.022
HbA1c  > 5.7%, n (%)	8 (7.84)	8 (28.57)	16 (12.31)	.007
HbA1c IFCC mmol/mol median (IQR)	34 (31, 36)	37 (33, 39)	35 (31, 37)	.017
HbA1c IFCC > 39 mmol/mol, n (%)	8 (7.84)	8 (28.57)	14 (10.77)	.007
Glucose mg/dL, median (IQR)	84 (78.5, 93)	89 (83.75, 98.5)	86 (79, 95)	.021
Glucose above 99 mg/dL, n (%)	8 (7.8)	7 (25)	15 (11.5)	.017
Insulin µU/mL, median (IQR)	6.7 (5.075, 9.2)	7.7 (5.7, 11.2)	6.9 (5.1, 10.1)	.37
QUICKY, median (IQR)	0.351 (0.33, 0.37)	0.36 (0.345, 0.381)	0.36 (0.34, 0.38)	.06
QUICKY < 0.34, n (%)	14 (15.2%)	8 (29.6%)	22 (18.5%)	.096
HOMA-IR, median (IQR)	1.413 (1.046, 2.0310	1.730 (1.284, 2.535)	1.44 (1.07, 2.17)	.309
HOMA-IR > 2, n (%)	23 (25)	10 (37)	33 (27.7)	.223
TyG index, median (IQR)	8.677 (8.446, 9.061)	8.675 (8.355, 8.934)	8.675 (98.372, 8.942)	.384
C-peptide before KTx (ng/mL), median (IQR)	5.9 (4.235, 9.4)	7.315 (6.035, 9.06)	6.695 (4.48, 9.38)	.06
Albumin before KTx (g/L) median (IQR)	39 (37, 41)	39 (37, 40)	3.9 (3.7, 4.1)	.56
Homocystein µmol/L, median (IQR)	22.2 (16.9, 31.7)	22.9 (18.575, 28.657)	22 (17.1, 31.4)	.847
Homocystein > 15 µmol/L	79 (77.5)	24 (85.7%)	103 (79.2)	.344
Cholesterol mg/dL, median (IQR)	198 (162.5, 229)	189 (160, 231.25)	197 (162, 229.00)	.647
TG mg/dL, median (IQR)	138 (100, 174.5)	133 (104.5, 174)	138 (100, 176)	.414
HDL mg/dL, median (IQR)	45 (38, 55)	47 (39, 56)	45.5 (38, 56)	.992
LDL mg/dL, median (IQR)	121.7 (92.1, 152.95)	104.7 (89, 142.5)	120 (91, 153)	.608
Uric acid mg/dL, median (IQR)	4.8 (3.7, 6.2)	4.2 (3.2, 5.3)	4.7 (3.5, 6.15)	.38

HOMA-IR = HOmeostatic Model Assesment-Insulin Resistance, HOMA-IR = fasting insulin [mU/mL] × fasting glucose [mg/dL]/405.

QUICKI (Quantitative Insulin Sensitivity Check Index) = 1/log fasting insulin [µU/mL] × log fasting glucose [mg/dL].

TyG index = ln [fasting triglycerides (mg/dL) × fasting glucose (mg/dL)/2].

HBA1c = glycosylated hemoglobin, Type A1c; HDL = high-density lipoprotein; HOMA-IR Index = homeostatic model assessment of insulin resistance index; KTx = kidney transplantation; N = number of patients; OR = odds ratio; QUICKI = quantitative insulin-sensitivity check index; TyG index = triglycerides and glucose index.

3.6.1.2. Fasting glucose

The median fasting glucose level measured before KTx was significantly higher in the group that later developed PTDM (P = .022). Moreover, IFG (defined as glucose > 99 mg/dL) was more frequently observed in the PTDM group: 21.4% of PTDM patients versus 8.4% of non-PTDM patients (P = .017; Table 2).

3.6.2. Statistically non-significant parameters

No significant differences were found between PTDM and non-PTDM patients in baseline levels of insulin, C-peptide, albumin, homocysteine, uric acid, lipid profile, or insulin resistance markers (HOMA-IR, QUICKI, TyG index). Data for the entire study group, as well as for PTDM and non-PTDM patients, are presented in Table 2.

3.6.3. Univariate analysis

In the univariate analysis, statistically significant risk factors for PTDM included older age, higher BMI, elevated pre-transplant HbA1c, and elevated pre-transplant fasting glucose (Table 3). HbA1c values above the normal range (≥5.7%/ ≥39 mmol/mol) were associated with a 4.45-fold increase in the risk of developing PTDM. Pre-transplant glucose levels above 99 mg/dL were associated with a 3.92-fold increased risk of PTDM (Table 3).

Table 3.

Univariate binary logistic regression of PTDM risk factors in KTx patients.

Variable	N	OR	95% CI	P-value
Dialysis modality before KTx	130
HD		–	–
PD		0.80	0.25, 2.26	.7
PREE		0.68	0.10, 2.83	.6
Gender	130
Female		–	–
Male		0.83	0.35, 1.96	.7
Age (yr)	130	1.04	1.01, 1.08	.019
BMI kg/m2	130	1.12	1.01, 1.25	.044
BMI ≥ 30	130
0		–	–
1		2.26	0.71, 6.63	.15
Number	130
1st		–	–
2nd and 3rd KTx		0.38	0.02, 2.17	.4
Pre-transplant dialysis time (mo)	128	0.99	0.98, 1.00	.3
Number of HLA mismatches	130	1.08	0.77, 1.52	.7
Cause of ESRD	130
GN		–	–
ADPKD		1.88	0.37, 7.73	.4
IN		0.83	0.12, 3.65	.8
Not known		1.00	0.14, 4.51	>.9
HN/IschN		2.92	0.91, 9.09	.065
Other (a/HUS)		3.00	0.55, 14.2	.2
Induction use (basiliximab/anti-thymocyte globulin)	130
0		–	–
1		2.00	0.78, 5.83	.2
Tacrolimus use (0), Cyclosporin use (1)	130
0		–	–
1		0.44	0.10, 1.40	.2
2-wk cumulative methylprednisolon dose	130	0.73	0.34, 1.46	.4
AR	130
0		–	–
1		1.88	0.25, 10.2	.5
DGF	130
0		–	–
1		0.57	0.16, 1.67	.3
CSC 1 mo after KTx (mg/dL)	130	0.91	0.48, 1.50	.7
eGFR 4p MDRD 1 mo after KTx (mL/min/1.73 m2)	130	1.00	0.98, 1.01	.9
Duration of hospitalization after KTx (days)	130	1.01	0.95, 1.06	.8
CMV infection	130
0		–	–
1		0.79	0.12, 3.33	.8
UTI	130
0		–	–
1		1.93	0.67, 5.23	.2
WIT (min)	129	1.03	0.98, 1.09	.2
CIT (min)	128	1.00	1.00, 1.00	.073
HbA1c before KTx (mmol/mol)	125	3.69	1.26, 12.0	.022
HbA1c > 5.7 mmol/mol	125
0		–	–
1		4.45	1.48, 13.5	.007
Glycemia before KTx (mg/dL)	123	1.04	1.01, 1.08	.021
Glycemia before KTx > 99 (mg/dL)	130
0		–	–
1		3.92	1.25, 12.1	.017
Insulin level before KTx (µU)	123	1.02	0.97, 1.07	.4
HOMA-IR Index before KTx	119	1.08	0.92, 1.27	.3
HOMA over the normal range before KTx > 2	119
0		–	–
1		1.76	0.69, 4.36	.2
QUICKI under the normal range before KTx < 0.34	119
0		–	–
1		2.16	0.77, 5.77	.13
TyG index	122	1.68	0.72, 4.04	.2
Homocysteine before KTx (µmol)	125	1.00	0.97, 1.03	.8
Homocysteine before KTx (µmol) over the normal range ≥ 15 µmol	130
0		–	–
1		1.75	0.60, 6.39	.3
Cholesterol before KTx (mg/dL)	127	1.00	0.99, 1.01	.6
Triglyceride before KTx (mg/dL)	127	1.00	1.00, 1.01	.4
HDL cholesterol before KTx (mg/dL)	126	1.00	0.97, 1.03	>.9
LDL cholesterol before KTx (mg/dL)	126	1.00	0.99, 1.01	.6
Uric acid before KTx (mg/dL)	124	0.89	0.69, 1.14	.4
C-peptide before KTx (ng/mL)	128	1.03	0.95, 1.11	.5
Albumin before KTx (g/L)	127	0.76	0.23, 2.39	.6

ADPKD = autosomal dominant polycystic kidney disease; AR = acute rejection; BMI = body mass index; CI = confidence interval; CIT = cold ischemia time; CSC = creatinine serum concentration; DGF = delayed graft function; eGFR CKD-EPI = estimated glomerular filtration rate and chronic kidney disease epidemiology collaboration; eGFR 4p MDRD = estimated glomerular filtration rate and 4-point modification of diet in renal disease formula; ESRD = end-stage renal disease; HBA1c = glycosylated hemoglobin, Type A1c; HD = hemodialysis; HDL = high-density lipoprotein; HLA = human leucocytes antigens; IschN = ischemic nephropathy; HOMA-IR Index = homeostatic model assessment of insulin resistance index; HUS = hemolytic uremic syndrome; KTx = kidney transplantation; N = number of patients; OR = odds ratio; PD = peritoneal dialysis; PREE = preemptive kidney transplantation; QUICKI = quantitative insulin-sensitivity check index; TyG index = triglycerides and glucose index; WIT = warm ischemia time.

3.6.4. Multivariate logistic regression model

Stepwise forward regression using the Bayesian Information Criterion identified age as the only significant predictor of PTDM. Similarly, the logistic regression model with elastic net regularization also identified age as the sole predictor.

4. Discussion

In our study, diabetes was diagnosed in 21.5% of patients. Additionally, 30 patients were diagnosed with prediabetes, and overall, 44.6% of patients exhibited glucose metabolism abnormalities. Comparable frequencies of glucose disturbances have been reported in previous studies. According to the literature, the use of OGTT has revealed a high prevalence of abnormal glucose metabolism in patients following KTx (KTx): 43% (11% with diabetes; 32% with IGT and/or IFG).^[22] Other studies have reported prevalence rates ranging from 32%^[20] to 51% (10% diabetes; 41% IGT and/or IFG).^[23]

As IFG and IGT are independent risk factors for the future development of PTDM,^[24] OGTT helps identify patients at risk and serves as a strong rationale for implementing non-pharmacological interventions to reduce the risk of diabetes. However, the most appropriate time to assess the risk of glycemic abnormalities is the pre-transplant period.

In our prospective cohort observational study, we did not have access to pre-transplant OGTT results, as this test was not routinely performed in most dialysis centers. Patients typically arrive at the transplant center shortly before the procedure, having fasted for at least 6 hours, and there is no logistical capacity to perform OGTT within that timeframe. Instead, we measured fasting glucose and HbA1c levels, which were both significantly associated with the development of PTDM.

In our cohort, HbA1c levels indicative of prediabetes were observed in 12.3% of patients and were associated with a 4.4-fold increased risk of PTDM. In the univariate analysis, statistically significant risk factors for PTDM included older age, higher BMI, elevated pre-transplant HbA1c, and elevated fasting glucose levels. Specifically, HbA1c values above the normal range (≥5.7%/ ≥39 mmol/mol) were associated with a 4.45-fold increase in the risk of PTDM. In the multivariate logistic regression model, age was identified as the sole independent predictor.

The measurement of HbA1c is useful in dialysis patients; however, it should be noted that HbA1c levels in this population tend to be underestimated, primarily due to the shortened lifespan of red blood cells, anemia, and erythropoietin therapy, which increases the proportion of young red blood cells with lower glycation levels. If possible, an OGTT should be performed during the pre-transplant period, as it is a more sensitive test and the diagnostic gold standard for evaluating candidates for KTx. However, OGTT requires considerable patient effort and is often impractical.

In situations where OGTT cannot be performed, HbA1c remains a valuable tool for identifying patients at risk of developing diabetes after transplantation. Another simple indicator associated with PTDM in our study was pre-transplant fasting glucose. The median fasting glucose level prior to transplantation was significantly higher in the group that later developed diabetes (P = .02). Glucose levels above 99 mg/dL were associated with a 3.92-fold increased risk of PTDM.

Four patients had both elevated HbA1c and IFG levels; 3 of them (75%) developed diabetes, while 1 did not. Notably, the individual who did not develop diabetes was neither overweight nor obese. In our cohort, a higher BMI was associated with an increased risk of PTDM.

When diagnosing glycemic disorders in the post-transplant period, it is important to note that the formal diagnosis of PTDM should be made no earlier than 3 months after KTx. This delay helps to avoid misclassifying transient hyperglycemia, which can occur in up to 90% of kidney transplant recipients during the early post-operative weeks.^[20,24]

Most patients experience an improvement in glucose tolerance as immunosuppressive therapy is tapered. However, the development of early post-transplant hyperglycemia remains a strong predictor of PTDM.^[20] In our study, 75% of patients with PTDM (21 out of 28) remained on intensive insulin therapy from the time of transplantation (initially for the management of post-transplant hyperglycemia, and later fulfilling the diagnostic criteria for PTDM).

In milder cases, the most effective method for diagnosing prediabetes and PTDM is the OGTT. Capillary blood glucose monitoring in the afternoon may serve as a more practical alternative for detecting hyperglycemia in clinical settings.^[1,25] Furthermore, the use of HbA1c is recommended for diagnosing PTDM,^[1] without the earlier concerns about its reliability in patients with end-stage renal disease (ESRD) or those on dialysis.

Tacrolimus is a more potent anti-rejection agent than cyclosporine A, but it has also been shown to have a greater diabetogenic effect.^[19,26,27] In our study, tacrolimus was part of the immunosuppressive regimen in 89% of PTDM patients and 78% of non-PTDM patients, although this difference was not statistically significant. The mean tacrolimus trough level 2 weeks after KTx was similar in both groups, approximately 12 ng/mL (Table 1). Likewise, there was no difference between groups in the cumulative dose of methylprednisolone per kilogram of body weight.

Induction therapy theoretically enables effective immunosuppression with lower levels of diabetogenic agents, such as calcineurin inhibitors and corticosteroids. In our cohort, the proportion of patients who received induction therapy did not differ significantly between those who developed PTDM and those who did not, likely because tacrolimus, cyclosporine, and steroid dosages were also comparable between the groups.

When selecting an immunosuppressive regimen, the risk of PTDM should be carefully balanced against the risk of AR for each individual patient. In general, immunosuppressive protocols that offer the best outcomes for both patient and graft survival should be prioritized, regardless of PTDM risk.^[10,19] PTDM or post-transplant hyperglycemia should be treated appropriately, independent of the immunosuppressive regimen used. In our study, the type of immunosuppressive therapy did not significantly influence the development of diabetes; however, general clinical markers such as BMI, glycated hemoglobin (HbA1c), and fasting glucose levels demonstrated a significant association.

PTDM is characterized by a combination of insulin resistance and insulin hyposecretion. However, pancreatic β-cell dysfunction is considered the key determinant of insulin hyposecretion and worsening glucose tolerance following renal transplantation.^[28] In our study, patients who developed PTDM did not exhibit higher insulin resistance, as assessed prior to transplantation using HOMA-IR, QUICKI, and TyG index. Similarly, pre-transplant insulin and C-peptide levels were comparable between the PTDM and non-PTDM groups.

There were no significant differences between PTDM and non-PTDM patients with respect to sex, cause of ESRD, method or duration of pre-transplant dialysis, number of HLA mismatches, ischemic time, immunosuppressive regimen, incidence of AR, delayed graft function, need for reoperation, or baseline laboratory parameters including lipid profile, insulin, C-peptide, homocysteine, uric acid, and albumin.

Post-transplant complications (such as UTI, CMV infection, surgical complications requiring reoperation, graft loss, and mortality) were not more frequent among patients with PTDM in our study.

Overall graft function during the observation period was good in both PTDM and non-PTDM groups. One month after KTx, the median serum creatinine concentration and eGFR (Modification of Diet in Renal Disease formula) did not differ significantly: in the PTDM group, 1.56 mg/dL and 59.6 mL/min; in the non-PTDM group, 1.605 mg/dL and 57.8 mL/min.

Our study has its limitations, including a relatively small study group, its single-center design, a limited duration of follow-up, and the lack of longitudinal assessment of metabolic parameters in the later post-transplant period. However, we believe that highlighting the significant importance of easily accessible laboratory markers, such as glucose and HbA1c measurements, represents an important conclusion of this prospective study.

5. Conclusion

Key risk factors for the development of PTDM after KTx include older age, higher BMI, and elevated pre-transplant serum glucose and HbA1c levels. In the studied cohort, general risk factors for diabetes appeared to play a more prominent role in the development of PTDM than transplantation-specific factors. Despite its limitations in patients with ESRD, HbA1c remains a useful tool for identifying individuals at increased risk of developing PTDM.

Acknowledgments

We would like to thank Mr. Adam Wyszomirski, M.Sc. Eng., for his invaluable assistance with the statistical analysis. His expertise significantly contributed to the integrity and interpretation of the data presented in this study.

Author contributions

Conceptualization: Beata Bzoma, Alicja Dębska-Ślizień.

Data curation: Beata Bzoma, Joanna Konopa, Andrzej Chamienia, Justyna Kostro.

Formal analysis: Beata Bzoma.

Funding acquisition: Alicja Dębska-Ślizień.

Investigation: Beata Bzoma.

Methodology: Beata Bzoma.

Project administration: Beata Bzoma.

Resources: Beata Bzoma.

Software: Beata Bzoma.

Validation: Beata Bzoma.

Visualization: Beata Bzoma.

Writing – original draft: Beata Bzoma.

Writing – review & editing: Alicja Dębska-Ślizień.