Should Early Hyperglycemia Be Considered a Risk Factor for Post-Transplant Diabetes Mellitus? Findings from a Retrospective Cohort Study in Kidney Transplant Recipients Without Diabetes Mellitus Prior to Transplant

Abstract

Background: Post-transplant diabetes mellitus (PTDM) is a complication of kidney transplantation, but the impact of early hyperglycemia (EH) remains unclear. This study aimed to assess the incidence of PTDM in kidney transplant recipients (KTRs) who experienced EH compared to those who do not at 6 months post-transplant. Methods: A single-center, retrospective cohort study was conducted in adults who underwent kidney transplantation from 1 January 2019 to 25 May 2022. KTRs who developed EH were compared against those who did not. Results: The primary outcome was the difference in incidence of PTDM at 6 months. Secondary outcomes included rehospitalizations and infections within 6 months and PTDM, renal function, cardiovascular events, and graft and patient survival within 12 months. Two hundred and seventy-nine KTRs (EH, n = 204 vs. comparator, n = 75) were included. There were higher incidences of PTDM in the EH group compared to the comparator group at 6 months (11% vs. 1.4%, p = 0.012) and 12 months post-transplant (18.5% vs. 5.5%, p = 0.007). KTRs with EH had 8.9 times greater odds of developing PTDM (OR 8.9; 95% 1.2–67.3, p = 0.03) at 6 months. There was no significant difference found in other secondary outcomes. Conclusions: KTRs with EH had an increased incidence of developing PTDM.

1. Introduction

Post-transplant diabetes mellitus (PTDM) is a common complication in kidney transplant recipients (KTRs), with around 20% developing PTDM [1,2,3,4,5]. There are several risk factors for the development of PTDM related to transplantation, including immunosuppressants, stress, inflammation, and infection [1]. Additionally, the development of PTDM in KTRs impacts clinical outcomes such as graft and patient survival, infection, and cardiovascular events [1,2]. PTDM is defined by international consensus as random blood glucose (RBG) > 200 mg/dL plus symptoms, fasting blood glucose (FBG) of >125 mg/dL, two-hour blood glucose after a 75 g oral glucose tolerance test of >200 mg/dL, or glycated hemoglobin (HbA1c) > 6.5% (48 mmol/mol) [6]. However, HbA1c as a sole diagnostic test should be used cautiously, as normal HbA1c may not accurately reflect glycemic control in the setting of anemia or changes in renal function in end-stage renal disease (ESRD) or post-transplant [6]. Rather, utilizing HbA1c in combination with blood glucose monitoring or an oral glucose tolerance test is recommended [6]. The international consensus meeting in 2013 established that a diagnosis of PTDM should exclude the first 45 days post-transplant [6]. Ideally, a formal diagnosis should occur once the KTR is stable on maintenance immunosuppression, free of acute infection, and with stable kidney allograft function [6]. The international consensus recognizes that early hyperglycemia (EH) occurring acutely and within 45 days of transplant is common and may reflect other acute processes such as infection, critical condition of the KTR, or immunosuppression [6]. KTRs, both with and without diabetes, can experience early hyperglycemia post-transplantation. EH has been defined as occurring within 45 days post-transplant and includes an inpatient FBG of >140 mg/dL, outpatient FBG > 126 mg/dL, RBG > 200 mg/dL, or the use of insulin after transplantation within 45 days post-transplant [6,7,8,9]. The prevalence of early hyperglycemia has been reported to be up to 90% among kidney transplant recipients during this period, but there is limited data assessing the long-term impact of early hyperglycemia in the acute post-transplantation period or if this is also a risk factor for developing PTDM [6,7].

Early hyperglycemia in KTRs can be due to multiple variables impacting their glucose levels, including the use of immunosuppressants, such as high-dose steroids, and changes in their renal function. Further, patients with ESRD may exhibit improved glucose control as exogenous insulin is primarily cleared renally, and uremia from ESRD can impact appetite and liver metabolism of endogenous insulin [7]. Following renal transplantation, kidney function improvement can increase insulin clearance, leading to worsening hyperglycemia [7].

There are several risk factors associated with the development of PTDM. These include obesity, older age, male sex, acute rejection, immunosuppressant use (namely that of corticosteroids and calcineurin inhibitors), hepatitis C virus (HCV), cytomegalovirus infection (CMV), human leukocyte antigen (HLA) mismatch, and deceased-donor kidney [8]. However, EH is not currently listed as a risk factor of PTDM [10]. Recent data suggests that EH in KTRs increases the risk of rehospitalizations due to infections, acute graft rejections, cardiovascular events, and PTDM [7,8,9,10,11].

The aim of this study is to assess the incidence of PTDM in KTRs without diabetes mellitus who experience EH compared to those who do not at six months post-transplant.

2. Materials and Methods

A single-center, retrospective cohort study was conducted in adult KTRs who underwent kidney transplantation from 1 January 2019 to 25 May 2022. This study (23-1299) was approved by the institution’s International Review Board on 6 July 2023 and was conducted in accordance with the Declaration of Helsinki. Informed consent was waived by the International Review Board as this study was retrospective. KTRs who developed early hyperglycemia (EH group) were compared against those who did not (comparator group). KTRs were followed for up to 12 months post-transplant. KTRs with an ICD-10 code for type 1 or type 2 diabetes mellitus prior to transplant were excluded. If no ICD-10 codes were present for diabetes mellitus, KTRs were then excluded if they had HbA1c ≥ 6.5% (48 mmol/mol) prior to transplant or on the day of transplant. KTRs with pre-diabetes, defined as baseline HbA1c of 5.7–6.4% (39–46 mmol/mol), were not excluded from the study. KTRs with an active ambulatory prescription for an antihyperglycemic agent prior to transplant or KTRs who received an ambulatory antihyperglycemic agent post-transplant for indications other than diabetes (i.e., heart failure, weight loss, renal protective effects) were excluded. Inpatient criteria were assessed for the index transplant admission only. Any subsequent hospitalizations within 45 days of kidney transplant were not assessed for EH as the acute events precipitating hospitalization could impact glycemic values.

2.1. Immunosuppression

KTRs received alemtuzumab 30 mg intravenous (IV) intra-operatively, anti-thymocyte globulin IV with a total dose of 6 mg/kg, or basiliximab 20 mg IV on post-operative day zero (POD0) and on POD3 or POD4, per institutional protocol based on immunologic risk and age at time of transplant. Methylprednisolone 500 mg IV intra-operatively was given and followed by 250 mg IV on POD1 with 125 mg IV on POD2 and POD3. Corticosteroids were continued with oral prednisone for patients who received basiliximab induction, were highly sensitized, or required chronic prednisone use prior to transplant. Otherwise, steroids were discontinued starting POD4. KTRs were started on immediate-release tacrolimus on POD0 and initially on mycophenolate mofetil 750 mg twice daily. The target tacrolimus trough concentrations were 8–10 ng/mL for the first three months, then 6–8 ng/mL for months 4 to 12. All tacrolimus concentrations were whole-blood samples. Maintenance immunosuppression data was collected at months 3, 6, and 12. Continuation of prednisone at discharge was also assessed. The tacrolimus coefficient of variation (CV) was calculated to estimate intrapatient variability using the equation CV (%) = (standard deviation/mean tacrolimus trough level) × 100.

2.2. Outcomes

The primary outcome was the difference in developing PTDM in KTRs without diabetes mellitus between the EH group and the comparator group at 6 months post-transplant. Secondary outcomes included PTDM rates within 12 months, rehospitalizations and infections within 6 months, and tacrolimus trough concentrations and CVs, prednisone use, acute rejection, graft and patient survival, renal function, and cardiovascular events within 12 months. All data points collected at 3, 6, and 12 months were assessed within 30 days of the timepoint, except for serum creatinine, which was assessed within 14 days. KTRs had to survive at least 6 months post-transplant to be assessed for the development of PTDM at 6 and 12 months.

2.3. Study Definitions

EH was defined as occurring within 45 days after transplant based on either inpatient or outpatient criteria. KTRs met inpatient criteria for EH if one of the following were met: two readings of random blood glucose ≥ 200 mg/dL, insulin use of ≥ 5 units within a 24 h period, or fasting blood glucose ≥ 140 mg/dL. KTRs met outpatient criteria for EH if one of the following were met: prescription of insulin at time of discharge, two readings of fasting blood glucose ≥ 126 mg/dL, or two readings of random blood glucose ≥ 200 mg/dL.

Inpatient glucose readings were considered a fasting blood glucose reading if obtained between 0000 and 0800. Outpatient glucose readings were considered a fasting blood glucose reading if obtained between 0000 and 1000. Any glucose levels obtained outside of these time frames were considered random blood glucose readings. Insulin used for hyperkalemia management was not included in the total units of insulin assessed within a 24 h period for EH eligibility.

PTDM was defined as occurring > 45 days post-transplant with the use of antihyperglycemic agent(s) or two of the following values: HbA1c ≥ 6.5% (48 mmol/mol), fasting blood glucose ≥ 126 mg/dL, or random blood glucose ≥ 200 mg/dL at 6- and 12-month follow-up appointments.

Incidences of bacteremia, urinary tract infections (UTIs), CMV DNAemia, and BK viremia within 6 months were analyzed. Bacteremia was defined as KTRs with 1 of 2 or 2 of 2 positive blood cultures who received treatment. UTI was defined as a urine culture with >10,000 colony-forming units (CFU) of bacteria and treatment with antimicrobial agents. CMV DNAemia was defined as a CMV PCR with a viral load > 1000 IU/mL within 6 months. BK viremia was defined as a BK PCR with a viral load > 250 IU/mL within 6 months. Additionally, the number of KTRs readmitted and the frequency of readmissions within 6 months were assessed. Serum creatinine was collected at months 3, 6, and 12 post-transplant. Serum creatinine was then used to calculate the estimated glomerular filtration rate with the Chronic Kidney Disease Epidemiology Collaboration 2021 (CKD-EPI 2021) equation. Cardiovascular events within 12 months were categorized as myocardial infarctions or transient ischemic attacks.

2.4. Statistical Analysis

For statistical analyses, primary and secondary outcomes were calculated using Chi square for categorical variables. Continuous variables including baseline characteristics were calculated using the t-test for normally distributed data and the Mann–Whitney U test for non-normally distributed data. A multivariable logistic regression model was conducted to assess the impact of other variables, including age (>45 years old), sex, race, induction agent, and presence of pre-diabetes, on PTDM incidence at 6 and 12 months. p-values of <0.05 were determined to be statistically significant.

3. Results

3.1. Patient Population

A total of 279 KTRs (n = 204 in the EH group vs. n = 75 in the comparator group) were included. The majority (95.1%) of KTRs met criteria for EH during their index admission, whereas 10 KTRs met only outpatient criteria for EH. All 10 KTRs who exclusively met outpatient inclusion criteria had two or more FBG values > 126 mg/dL. A majority of KTRs in the EH group met inpatient criteria for elevated fasting blood glucose values (n = 160) or insulin use (n = 116) compared to those within the EH group who met random blood glucose > 200 mg/dL on two or more occasions (n = 67). All three inpatient criteria for EH were met for 42 KTRs.

3.2. Baseline Characteristics

Baseline characteristics (

Table 1. Baseline characteristics.

	EH (N = 204)	Comparator (N = 75)	p-Value
Sex, male, n (%)	119 (58.3)	29 (38.7)	0.004
Age at time of transplant, years; median (IQR)	51 (37–62)	47 (38.5–55)	0.092
Race, n (%)
Black	70 (34.3)	44 (58.6)	<0.005
White	94 (46.1)	20 (26.7)	0.004
Other	40 (19.6)	11 (14.7)	0.344
Body mass index (kg/m2), mean (SD)	28.1 (5.5)	28.7 (5.6)	0.446
Etiology of kidney disease, n (%)
Hypertensive nephrosclerosis	52 (25.5)	26 (34.7)	0.13
FSGS	23 (11.3)	11 (14.6)	0.443
Polycystic kidney disease	24 (11.7)	5 (6.7)	0.216
IgA nephropathy	13 (6.4)	7 (9.3)	0.395
Other	92 (45.1)	26 (34.7)	0.118
cPRA, % median (IQR), [min, max]	0 [0, 100]	0 [0, 63]	0.495
cPRA 0–20, n (%)	196 (96.1)	73 (97.3)	0.617
cPRA 21–60, n (%)	3 (1.5)	1 (1.3)	1
cPRA 61–100, n (%)	5 (2.4)	1 (1.3)	1
Delayed graft function, n (%)	42 (20.6)	11 (14.7)	0.264
KDPI, % for deceased donors, median (IQR)	38 (33–44)	30 (15.3–40.3)	0.081
Donor type, n (%)
DBD	99 (48.5)	42 (56)	0.269
DCD	53 (26)	14 (18.7)	0.205
Living	52 (25.5)	19 (25.3)	0.979
Cold ischemia time for deceased donors (hours), mean (SD)	15.4 (5.75)	13.9 (5.7)	0.095
Induction agent, n (%)
ALZ	111 (54.4)	51 (68)	0.041
ATG	79 (38.7)	20 (26.7)	0.062
Basiliximab	14 (6.9)	4 (5.3)	0.645
HCV donor status, n (%)
HCV Ab + NAT+	27 (13.2)	6 (8)	0.23
HCV Ab + NAT−	3 (1.5)	2 (2.7)	0.613
Positive HCV in recipient	9 (4.4)	1 (1.3)	0.297
Baseline HbA1c, % mean (SD)	4.97 (0.48)	4.82 (0.38)	0.018
Baseline HbA1c in pre-diabetes range (5.7–6.4%), n (%)	15 (7.4)	0 (0)	0.014
Length of stay, days (mean, SD)	7.3 (6)	6 (2.1)	0.074

ALZ: alemtuzumab; ATG: anti-thymocyte globulin; cPRA: calculated panel reactive antibodies; DBD: donation after brain death; DCD: donation after cardiac death; FSGS: focal segmental glomerulosclerosis; HbA1c: hemoglobin A1c; HCV: hepatitis C virus; KDPI: kidney donor profile index; SD: standard deviation; Ab: antibody; NAT: nucleic acid amplification testing.

) were similar between the groups, except for male sex (58.3% EH vs. 38.7% comparator, p = 0.004), Black race (34.3% vs. 58.6%, p < 0.005), baseline HbA1c (4.97 ± 0.48% vs. 4.82 ± 0.38%, p = 0.018), pre-diabetes (7.4% vs. 0%, p = 0.014), and alemtuzumab induction (54.4% vs. 68%, p = 0.041).

3.3. Incidence of PTDM

There was a significantly higher incidence of PTDM in the EH group compared to the comparator group at 6 months (11% vs. 1.4%, p = 0.012). KTRs with EH had 8.9 times greater odds of developing PTDM (OR 8.9; 95% CI 1.2–67.3, p = 0.03) at 6 months. For every 10 KTRs with EH, 1 additional patient would have PTDM at 6 months. The incidence of PTDM at 12 months post-transplant was also significantly higher in the EH group (18.5% vs. 5.5%, p = 0.007). KTRs with EH had 3.9 times greater odds of developing PTDM (OR 3.9; 95% CI 1.3–11.4, p = 0.01) at 12 months. For every 7 KTRs with EH, 1 additional patient would have PTDM at 12 months. The cumulative incidences of PTDM at 6 and 12 months for the EH and comparator groups are shown in Figure 1. Confounding variables such as gender (6 months 95% CI, −1.96 to 0.195, p = 0.108; 12 months 95% CI −0.888 to 0.645, p = 0.756), race (6 months 95% CI, −1.03 to 0.898, p = 0.889; 12 months 95% CI −0.97 to 0.54, p = 0.577), induction agent (6 months 95% CI, −0.912 to 0.975, p = 0.948; 12 months 95% CI −0.499 to 0.979, p = 0.525), and pre-diabetes (6 months 95% CI, −2.20 to 0.685, p = 0.303; 12 months 95% CI −1.9 to 0.642, p = 0.332) prior to transplant were not significant for impacting PTDM at 6 months or within 12 months. Within the comparator group, the KTRs who met criteria for PTDM at 6 and 12 months were all due to a prescription for an antihyperglycemic agent. Similarly, in the EH group, a majority of KTRs met criteria for PTDM due to an antihyperglycemic agent or had an antihyperglycemic prescription in addition to positive laboratory values (

Table 2. Criteria met for post-transplant diabetes mellitus (PTDM *) diagnosis in early hyperglycemia group.

Combinations Meeting Criteria in EH Group	PTDM * at 6 Months (N = 22)	Additional PTDM * from 6 Months to 12 Months (N = 15)
HbA1c AND FBG, n (%)	3 (13.6)	1 (6.7)
HbA1c AND RBG, n (%)	0 (0)	0 (0)
RBG AND FBG, n (%)	3 (13.6)	0 (0)
HbA1c AND RBG AND FBG, n (%)	1 (4.5)	0 (0)
Antihyperglycemic agent, n (%)	8 (36.4)	12 (80)
HbA1c AND FBG AND antihyperglycemic agent, n (%)	1 (4.5)	0 (0)
HbA1c AND RBG AND antihyperglycemic agent, n (%)	1 (4.5)	1 (6.7)
RBG AND FBG AND antihyperglycemic agent, n (%)	3 (13.6)	0 (0)
HbA1c AND RBG AND FBG AND antihyperglycemic agent, n (%)	2 (9.1)	1 (6.7)

HbA1c: hemoglobin A1c; FBG: fasting blood glucose; RBG: random blood glucose. * PTDM defined as prescription for antihyperglycemic agent or 2 of the following values: HbA1c ≥ 6.5% (48 mmol/mol), FBG ≥ 126 mg/dL, RBG ≥ 200 mg/dL.

).

3.4. Secondary Outcomes

Secondary outcomes are shown in

Table 3. Secondary outcomes.

	EH (N = 204)	Comparator (N = 75)	p-Value
Tacrolimus trough concentration, ng/mL, mean (SD)
0–3 months 3–6 months 6–12 months	8.1 (1.1) 7.7 (1.3) 7.6 (1.7)	8.1 (1.4) 7.6 (1.6) 6.9 (1.8)	0.593 0.301 0.002
Tacrolimus coefficient of variation, mean (SD)
0–3 months 3–6 months 6–12 months	38.3 (10.7) 29.3 (15) 28.4 (16)	37.1 (11) 30.2 (17.9) 29 (14.7)	0.412 0.654 0.768
Number of recipients taking prednisone, n (%)
Discharge 3 months 6 months 12 months	39 (19.1) 67 (32.8) 77 (37.7) 76 (37.2)	12 (16) 19 (25.3) 29 (38.7) 31 (41.3)	0.55 0.229 0.888 0.535
Prednisone dose, mg, mean (SD)
3 months 6 months 12 months	8.8 (12) 6.6 (6.6) 5.8 (4)	6 (3.6) 6.9 (5.5) 6 (3.9)	0.334 0.841 0.707
Mycophenolate mofetil mg/day, median (IQR)
3 months 6 months 12 months	1500 (1000–1500) 1500 (1000–1500) 1500 (1000–1500)	1500 (1000–1500) 1000 (500–1500) 1000 (500–1500)	0.07 0.194 0.567
Renal function *, ml/min/1.73 m2, mean (SD)
3 months 6 months 12 months	57.2 (21.3) 58.3 (21.7) 59.1 (22.6)	58.8 (23) 61.4 (19.2) 62.3 (20)	0.603 0.285 0.337
Infections within 6 months, n (%)
Bacteremia UTI Recipients with ≥2 UTIs Cytomegalovirus DNAemia BK viremia	11 (5.4) 33 (16.2) 14 (6.9) 13 (6.4) 26 (12.7)	1 (1.3) 12 (16) 6 (8) 7 (9.3) 10 (13.3)	0.191 0.972 0.795 0.434 0.897
Readmissions within 6 months, n (%)	120 (58.8)	40 (53.3)	0.411
Number of readmissions within 6 months, n (%)
1 2 3 ≥4	65 (54.2) 30 (25) 12 (10) 13 (10.8)	20 (50) 11 (27.5) 4 (10) 5 (12.5)	0.647 0.754 1 0.776
Cardiovascular events within 12 months, n (%)	2 (1)	1 (1.3)	1
Myocardial infarction Transient ischemic attack	1 1	0 1
Biopsy-proven acute rejection within 12 months, n (%)	15 (7.4)	8 (10.7)	0.461
Graft survival within 12 months, n (%)	204 (100)	75 (100)	1
Patient survival within 12 months, n (%)	199 (97.5)	72 (96)	1

UTI: urinary tract infection. * Calculated using the CKD-EPI creatinine equation (2021) gender.

. Of the secondary outcomes, mean tacrolimus trough concentrations between six and twelve months post-transplant were significantly different in the EH group compared to the comparator group (7.6 ± 1.7 vs. 6.9 ± 1.8 ng/mL, p = 0.002). A multivariable logistic regression conducted with all participants did not identify any variables significant at 6 months but did identify age > 45 years old as being significant for developing PTDM at 12 months (1.98, 95% CI 0.009 to 1.67, p = 0.048). Fifteen KTRs with EH had pre-diabetes prior to transplant. There were no KTRs in the comparator group with pre-diabetes. Of the 15 KTRs with pre-diabetes in the EH group, 4 developed PTDM (26.7%). In a subgroup analysis of KTRs with pre-diabetes, baseline characteristics differed in sex (male 94%), race (Black 66.67%), and average tacrolimus concentration and coefficient of variation at 3 months (7.7 ng/mL + 1.81 and 43.7 + 16.31), 6 months (8.2 ng/mL + 1.74 and 36.2 + 28.26), and 12 months (7.4 ng/mL + 1.15 and 25.8 + 7.24). In a sub-group analysis of KTRs without pre-diabetes, a multivariable logistic regression did not identify any variables significant at 6 months but identified age > 45 years old as being significant for PTDM at 12 months (2.20, 95% CI 0.12 to 2.1, p = 0.028).

Prescription for outpatient antihyperglycemic agents > 45 days post-transplant was assessed, and the time to antihyperglycemic agent for those with PTDM in the EH group is shown in Figure 2. For those who received antihyperglycemic prescriptions in the EH group, the most commonly prescribed medication included metformin (59%), insulin (9%), glucagon-like peptide 1 receptor agonists (9%), or combination therapy with metformin and either a glucagon-like peptide 1 receptor agonist or a dipeptidyl peptidase 4 inhibitor (9%). The average time from transplant to antihyperglycemic agent was 273 days in the comparator group and 201.7 days in the EH group (p = 0.227).

4. Discussion

This study observed a significant difference in the incidence of PTDM at 6 and 12 months in KTRs without diabetes mellitus at baseline who experienced EH post-transplant compared to those who did not. This adds to the current literature that characterizes the potential impact of early hyperglycemia and the risk of developing PTDM. In a retrospective cohort analysis conducted by Chakkera and colleagues, a similar relationship was seen, with KTRs who experienced acute hyperglycemia while inpatients following transplantation having a higher incidence of PTDM within the first 12 months post-transplant (29% in those with inpatient hyperglycemia vs. 4% in those without inpatient hyperglycemia). This study further elucidated that KTRs with high usage of insulin (>20 units of insulin during hospitalization) to control inpatient hyperglycemia were associated with a 2.4 higher risk of developing PTDM [12]. Additional retrospective analyses have found this association between early hyperglycemia, occurring during the first week post-transplant, and the development of PTDM [9,13]. Our study expands on the current literature by considering early hyperglycemia both inpatient and outpatient and up to 45 days post-transplant.

With limited studies assessing the impact of post-transplant EH on clinical outcomes, our study provides additional data to support the association of EH with the development of PTDM. While there are several transplant-specific risk factors identified for the development of PTDM, EH is currently not recognized as a risk factor [1,6]. Our findings confirm that EH in KTRs can increase the likelihood of having PTDM within the first 12 months post-transplant. Therefore, EH should be recognized and monitored in addition to other previously described risk factors for PTDM, including immunosuppression, acute rejection, inflammation from infections such as HCV and CMV, and deceased-donor kidneys [1]. In contrast to other studies, a difference in clinical outcomes such as rehospitalizations, infections, acute graft rejection, and cardiovascular events was not found [8,9,11,14]. The lack of difference in these secondary outcomes is likely due to the small sample size and short follow-up of our study. Previous studies which observed a difference in clinical outcomes included greater than 1000 KTRs with a follow-up period of 3–7 years [8,9,11,14].

Within the EH group, mean tacrolimus trough concentrations between 6 and 12 months were significantly higher compared to those in the comparator group. With calcineurin inhibitors established as a known risk factor of PTDM, this difference in tacrolimus trough concentrations represents a potential confounding variable when considering the development of PTDM at 12 months [1]. However, the incidence of PTDM was higher in the EH group compared to the comparator at 6 months, suggesting that EH contributed to the risk of PTDM development. Additionally, KTRs had higher average baseline HbA1c, which was significantly different from the baseline HbA1c of the comparator group, which may have contributed to the development of PTDM. However, the multivariable logistic regression model constructed from the baseline characteristics that differed significantly between groups did not identify this variable to be statistically significant in impacting the development of PTDM. The incidence of pre-diabetes at baseline was also higher in the EH group compared to the comparator group, which may suggest that several KTRs in the EH group had impaired glucose tolerance or dysglycemia prior to transplantation. Patients with ESRD may have decreased insulin clearance, decreased appetite due to uremia, and decreased liver metabolism of endogenous insulin. With changes in renal function during the transition from the native kidney to the allograft, there may be an increase in insulin clearance due to improved renal function, changes in appetite due to increased uremic toxin clearance, and improved liver metabolism of insulin, which may all contribute to impaired glycemic control post-transplant [7]. While HbA1c was collected at baseline and throughout the follow-up period to assess PTDM incidence, there are limitations to utilizing HbA1c in patients with ESKD or post-transplant. Given the complexities and limitations of HbA1c in these patient populations, it is recommended to not solely depend on HbA1c as a diagnostic factor or assessment of glycemic control. HbA1c may underestimate glycemic control in post-transplant patients and have lower sensitivity compared to utilizing an oral glucose tolerance test [15]. As such, this study did not rely on HbA1c alone to identify PTDM but instead required either a prescription for an antihyperglycemic agent or two or more positive laboratory findings (FBG > 126 mg/dL, RBG > 200 mg/dL, or HbA1c > 6.5% [48 mmol/mol]). Utilizing FBG in addition to HbA1c has been shown to increase the sensitivity of detecting PTDM compared to HbA1c alone [15]. Additionally, patients with ESRD may have falsely low HbA1c readings in the setting of anemia, while using erythropoietin-stimulating agents, or while on dialysis secondary to erythrocyte lysis [16]. These conditions may have impacted our patient population’s baseline HbA1c at the time of transplant and may not have accurately excluded those with a history of T2DM prior to transplant based on HbA1c. However, our study excluded patients with a historical diagnosis of type 1 or type 2 diabetes mellitus based on ICD-10 codes, and if no ICD-10 codes were present, then HbA1c levels at the time of transplant and up to 180 days prior to transplant were assessed and excluded if >6.5% (48 mmol/mol). Thus, there is a risk a patient included in the study could have had a historical HbA1c level > 6.5% (48 mmol/mol) without a documented ICD-10 code for diabetes mellitus, representing a limitation of this study.

There were a few limitations to this study. EH was broadly defined in order to include more KTRs, but a narrower definition in future studies may highlight KTRs at greater risk for PTDM. Additionally, this study did not assess the impact of tighter glucose control immediately post-transplant or within 45 days post-transplant on PTDM. Due to a smaller sample size in this study and in each cohort, this increased the risk of type two error. This limited sample size may explain why most secondary outcomes were not statistically significant. Additionally, the limited number of outcome events, particularly within the comparator group, limits the precision of our effect estimates. Thus, this limits the certainty of the association between EH and the development of PTDM. Due to the nature of retrospective chart review, there was an inability to ensure that morning glucose readings were accurately collected in a fasting state, especially in the outpatient setting. KTRs are often encouraged to maintain adequate oral hydration, which may include fluids containing carbohydrates, representing a limitation of this study when assessing fasting blood glucose readings. This study specifically identified when antihyperglycemic agents were first prescribed, but it is uncertain whether these medications were initiated by the patient or whether these therapies were continued beyond the first prescription. Weight gain leading to obesity is a known risk factor of PTDM [1]. The baseline weight and BMI for KTRs were included in our study, but weight, BMI, and change from baseline in weight at 6 and 12 months were not included, thus representing a limitation of whether KTRs who developed PTDM had a difference in weight gain or development of obesity post-transplant. Lastly, the follow-up period for this study was limited to 12 months.

These findings call to attention the importance of developing approaches to appropriately assess for EH in the early post-transplant period and of continued evaluation for the development of PTDM in the outpatient setting. In a small study assessing the impact on glucose control in the acute post-transplantation period, investigators found no difference in readmissions, hypoglycemia, or average FBG in KTRs treated with oral antihyperglycemic agents compared to insulin [17]. Future studies examining the impact of tighter glucose control in KTRs with EH should be conducted to determine whether interventions should be implemented to mitigate EH and its long-term effects on PTDM development. Considering the EH group had higher average baseline HbA1c and higher incidence of pre-diabetes, this may highlight the need to judiciously monitor KTRs with elevated baseline HbA1c more closely for EH and provide early interventions for glucose control.

5. Conclusions

This study demonstrated the incidence of EH and the development of PTDM in KTRs without diabetes mellitus at baseline. Given the small sample size and low incidence of PTDM, further studies should be conducted to fully elucidate the association between EH and PTDM. Given that PTDM and post-transplant complications develop over years, additional studies with extended follow-up periods are necessary. However, the association between EH and the development of PTDM within one year post-transplant in this study adds to the justification for EH to be considered a risk factor for PTDM.