Impact of specialised endocrinology care on metabolic control and healthcare utilisation outcomes after kidney transplantation in patients with diabetes: A 12‐month observational cohort study

Abstract

Aims

To evaluate whether an endocrinology‐integrated transplant clinic and differing healthcare delivery models are associated with metabolic outcomes during the first year after kidney transplantation in recipients with pre‐existing diabetes.

Materials and Methods

We conducted a retrospective longitudinal cohort study of adult kidney transplant recipients with diabetes at a US and a European academic centre. Participants were classified by post‐transplant diabetes care model: Cohort 1, endocrinology‐led Endocrine Transplant Clinic (ETC; n = 99); Cohort 2, historical standard transplant care at the same US centre (n = 81); and Cohort 3, standard endocrinology care at a Spanish academic centre (n = 40). Pre‐specified outcomes included HbA1c, body mass index (BMI), blood pressure, and lipid levels measured at baseline and 3, 6, and 12 months. Linear mixed‐effects models adjusted for demographic and clinical covariates were applied. Missing longitudinal data were addressed using multiple imputation with complete‐case sensitivity analyses.

Results

Among 220 recipients, adjusted metabolic trajectories were broadly similar across cohorts. HbA1c was unchanged at 3 and 6 months but higher at 12 months; >50% had HbA1c >7% at 1 year. BMI remained stable, with ≥30% meeting obesity criteria throughout follow‐up. Blood pressure did not improve, and systolic hypertension (>130 mmHg) remained common (49%–77%). At 12 months, LDL‐C ≥70 mg/dL was present in 20.8%, 63.3%, and 42.3% of Cohorts 1–3. Findings were consistent in sensitivity analyses.

Conclusions

Metabolic control in the first post‐transplant year showed stabilisation rather than improvement, with many recipients above cardiometabolic targets. Prospective studies should test whether earlier, protocolised multidisciplinary management improves cardiovascular and graft outcomes.

1. INTRODUCTION

Diabetes mellitus (DM) is a leading cause of end‐stage renal disease (ESRD) worldwide. In the United States, DM accounts for nearly half of all treated ESRD cases, with a prevalence of 2254 per million people in 2020. ¹ Kidney transplantation remains the treatment of choice for eligible patients with diabetic nephropathy, offering superior survival and quality of life compared to dialysis. ² The United States performs the highest number of kidney transplants globally, ² and timely, specialised post‐transplant care is essential for long‐term graft function and patient outcomes.^{3, 4, 5}

Post‐transplant diabetes management presents significant clinical challenges. Post‐transplant weight gain and the diabetogenic effects of immunosuppressive therapy complicate glycaemic control, and evidence to guide pharmacologic strategies remains limited, necessitating careful therapeutic selection. In the setting of poorly controlled diabetes, hyperglycaemia is associated with prolonged hospital stays, ⁶ higher readmission rates,^{7, 8} and increased use of emergency care. ⁹ Although patients receive intensive monitoring immediately after surgery, transplant teams may lack the endocrinology expertise required for optimal diabetes management of complex cases.

In late 2022, our institution established an Endocrine Transplant Clinic (ETC) within the Kidney Transplant Program. The ETC integrates endocrinology‐led care within the transplant framework to improve post‐transplant metabolic outcomes in patients with DM. Before its implementation, diabetes management after transplantation was fragmented, with referrals to general endocrinology clinics after hospital discharge and potential delays to specialist follow‐up. Consequently, early post‐transplant glycaemic management was frequently overseen by nephrologists, with endocrinology input provided on an ad hoc basis.

The ETC model provides a structured, multidisciplinary approach led by endocrinologists dual‐certified in Internal Medicine and Endocrinology, Diabetes, and Metabolism. The team includes a Certified Diabetes Care and Education Specialist nurse practitioner, pharmacists, and dietitians with renal and post‐transplant expertise to support coordinated metabolic care.

The comparative outcomes of an integrated endocrine‐transplant model versus usual care during the first post‐kidney transplant year remain uncertain, and cross‐system comparisons are scarce. We conducted a 12‐month retrospective longitudinal cohort study of adult kidney transplant recipients with pre‐existing diabetes comparing three models: (1) patients managed in the ETC, (2) patients previously treated at the same US centre before ETC implementation under standard transplant clinic care, and (3) recipients at a European academic centre within a different healthcare delivery system and structured endocrinology follow‐up after the early post‐transplant period. We hypothesised that care integration and healthcare context would be associated with differences in glycaemic control and cardiometabolic trajectories over 12 months. Accordingly, we evaluated metabolic outcomes and compared post‐transplant healthcare utilisation across care models and settings.

2. MATERIALS AND METHODS

2.1. Study design

We conducted a multicentre retrospective longitudinal cohort study of adults (≥18 years) with pre‐existing diabetes mellitus undergoing kidney transplantation to evaluate metabolic outcomes and healthcare utilisation over the first 12 months post‐transplant. All eligible recipients during the study periods were included. Recipients were eligible if they underwent kidney transplantation at participating centres and had at least two post‐transplant metabolic assessments recorded. Patients with new‐onset diabetes after transplant were excluded.

The study was approved by the Institutional Review Board at UC Davis Health (UCDH; approval code 2186298‐1) and by the Red Gallega de Comités de Ética de la Investigación (approval code 2022/163). All procedures were conducted in accordance with the Declaration of Helsinki.

2.2. Participants

A total of 220 adult kidney transplant recipients with DM were included (Table 1) and classified according to post‐transplant diabetes care model and centre into three cohorts: Cohort 1 (n = 99), recipients followed at a US university medical centre within an ETC model (1 October 2022–1 October 2024); Cohort 2 (n = 81), recipients transplanted at the same US centre before ETC implementation and followed in a transplant kidney clinic where diabetes care was delivered by non‐endocrinology clinicians; and Cohort 3 (n = 40), recipients transplanted at a Spanish academic centre between 2019 and 2024 within the national healthcare system. In Cohort 3, early post‐transplant follow‐up during the first 3 months was managed by transplant nephrology, after which patients resumed outpatient endocrinology follow‐up consistent with pre‐transplant care routines. No matching procedures were performed due to the limited number of eligible patients.

TABLE 1.

Clinical and metabolic characteristics at the time of kidney transplantation, by cohort.

	Cohort 1 (n = 99)	Cohort 2 (n = 81)	Cohort 3 (n = 40)	p
Age, years	61 (40–76)	61 (30–77)	63 (35–76)	NS
Sex, n (%), men	65 (65.6)	61 (75.3)	30 (75.0)	NS
DM duration, years	24.0 ± 9.2	24.4 ± 8.2	21.0 ± 10.9	NS
Hospital stays, days	3 (2–74)	4 (2–12)	15 (7–31)	<0.0001
HbA1c, % (mmol/mol)	6.8 ± 1.2 (51 ± 13)	6.7 ± 1.1 (50 ± 12)	7.0 ± 1.2 (53 ± 13)	NS
BMI, kg/m2	29.2 ± 4.9	28.3 ± 4.3	30.0 ± 4.2	NS
SBP, mmHg	137 ± 23 (87–185)	142 ± 20 (94–178)	146 ± 20 (110–187)	NS
DBP, mmHg	69 ± 14 (35–99)	71 ± 13 (45–108)	73 ± 11 (49–98)	NS
Total cholesterol, mg/dL	144 ± 42	135 ± 40	145 ± 50	NS
LDL‐C, mg/dL	67 ± 29	63 ± 32	66 ± 28	NS
HDL‐C, mg/dL	45 ± 16	43 ± 16	39 ± 9	NS
TG, mg/dL	172 ± 131	171 ± 140	185 ± 130	NS
Diabetic retinopathy, %
Non‐proliferative	30.4	43.1	38.5	NS
Proliferative	57.1	29.3	38.5	NS
Macular oedema	20.0	25.9	24.3	NS
CAD, %	32.3	41.1	27.5	NS
CAS, %	10.8	17.8	10.0	NS
MI, %	4.3	10.0	3.0	NS
CABG, %	11.8	8.2	0.0	NS
HFrEF, %	10.0	5.0	5.0	NS
HFpEF, %	9.0	10.0	12.5	NS
LEA, %	12.0	3.0	5.0	NS

Note: Cohort 1: Kidney transplant recipients receiving post‐transplant care at a US university medical centre through a specialised Endocrine Transplant Clinic (ETC) model. Cohort 2: Kidney transplant recipients managed at the same US academic centre before the ETC model was established. Cohort 3: Kidney transplant recipients transplanted and followed at a Spanish academic centre operating within the national healthcare system. Data are expressed as n (%), mean ± standard deviation, or median (total range) as appropriate. p‐value reflects between‐cohort comparisons.

Abbreviations: BMI, body mass index; CABG, coronary artery bypass grafting; CAD, coronary artery disease; CAS, coronary artery stent; DBP, diastolic blood pressure; DM, diabetes mellitus; HbA1c, glycated haemoglobin; HDL‐C, high‐density lipoprotein cholesterol; HFrEF, heart failure with reduced ejection fraction; HFpEF, heart failure with preserved ejection fraction; LDL‐C, low‐density lipoprotein cholesterol; LEA, lower extremities amputation; MI, myocardial infarction; SBP, systolic blood pressure; TG, triglycerides.

2.3. Variables and data sources

Clinical and biochemical data were extracted from electronic medical records from the transplant hospitalisation and scheduled outpatient follow‐up visits, at baseline (first post‐transplant visit), 3, 6, and 12 months. Measurements reflected routine clinical care.

Outcomes included:

• Metabolic and cardiovascular outcomes: BMI (kg/m²), HbA1c (%), systolic and diastolic blood pressure (SBP and DBP, mmHg), and fasting lipid profile (total cholesterol, LDL‐C, HDL‐C, and triglycerides [TG, mg/dL]).

• Healthcare utilisation outcomes: length of transplant hospitalisation and 30‐day hospital readmission.

To reduce confounding, covariates were pre‐specified based on clinical relevance and included age, sex, diabetes duration, dialysis duration, transplant hospitalisation length, and steroid exposure.

2.4. Missing data

Missing HbA1c values accounted for 30.2% of scheduled measurements. Missingness increased over time, from 3.18% at baseline to 21.4% at 3 months, 44.5% at 6 months, and 51.8% at 12 months, and varied across cohorts (26.0% in Cohort 1, 34.6% in Cohort 2, and 31.9% in Cohort 3). For other longitudinal variables (BMI, SBP, DBP, Total Cholesterol, HDL‐C, LDL‐C, and TG), missingness showed a similar monotonic increase over time, consistent with loss to follow‐up and intermittent nonattendance. Baseline covariates were nearly complete (<0.9% missing).

2.5. Statistical analysis

Baseline characteristics were summarised descriptively using observed data. Between‐cohort differences were evaluated using the Kruskal–Wallis and the chi‐square test.

Longitudinal outcomes were analysed using linear mixed‐effects models with fixed effects for cohort, time, and the cohort‐by‐time interaction, adjusting for pre‐specified covariates. Continuous covariates were standardised (z‐scores) to facilitate model convergence and coefficient comparability. A random intercept accounted for within‐subject correlation. Pairwise comparisons used Tukey adjustment.

Missing outcome were addressed using multiple imputation by chained equations under a missing‐at‐random assumption (20 datasets combined via Rubin's rules). Complete‐case analyses served as sensitivity analysis.

Adjusted marginal means and 95% CIs were estimated using the emmeans framework. Simulation‐based power analyses were performed for key interaction effects.

All analyses were conducted in R with a two‐sided significance at p‐value <0.05.

3. RESULTS

3.1. Baseline characteristics

Table 1 summarises cohorts composition and the variables analysed at the time of transplantation. Cohort 1 included 99 patients (65 men, 34 women); Cohort 2 included 81 (61 men, 20 women); and Cohort 3 included 40 (30 men, 10 women).

The racial and ethnic distribution of the US cohorts was 30% Hispanic, 26% Asian, 16% non‐Hispanic White, 13% African American, and 15% other or unknown. Cohort 3 consisted of Spanish White participants, with one Black African participant.

Although no formal matching was performed due to the limited eligible sample size, cohorts were broadly comparable at transplantation. Aside from racial and ethnic distribution, groups were similar in age, sex, diabetes duration, BMI, HbA1c, blood pressure, and lipid profile. Microvascular and macrovascular complications at transplantation are summarised in Table 1.

3.2. Healthcare utilisation

Time from dialysis initiation to transplantation was longer in the US cohorts (1715 ± 1194 days in Cohort 1 and 1746 ± 1074 in Cohort 2) than in Cohort 3 (782 ± 492 days; p < 0.001).

Transplant hospitalisation was longer in Cohort 3 (median 15 days) than in Cohorts 1 and 2 (median 3 and 4 days). Thirty‐day readmission rates were similar across cohorts (21.7%, 23.4%, and 22.5%).

3.3. Metabolic outcomes

3.3.1. Haemoglobin A1c

In the adjusted linear mixed‐effects model using imputed datasets, HbA1c did not differ from baseline at 3 or 6 months but was higher at 12 months (Figure 1A; Table S1). Baseline HbA1c was similar across cohorts, and the cohort‐by‐time interaction was not significant.

FIGURE 1.

Longitudinal changes in haemoglobin A1c (HbA1c) after kidney transplantation. (A) Adjusted mean HbA1c (95% CI) at baseline, 3, 6, and 12 months by cohort estimated from linear mixed‐effects models. HbA1c values at 3 and 6 months were similar to baseline, with higher values at 12 months. Trajectories were generally similar across cohorts, with a between‐cohort difference at 3 months. Within‐cohort increases at 12 months were observed in Cohorts 1 and 2 only. †p < 0.01 versus baseline; ‡p < 0.01 versus 3 months. *p < 0.05 between cohorts at the same time point. (B) Proportion of participants with HbA1c >7% at each visit by cohort.

Between‐cohort differences were generally not significant, except at 3 months when HbA1c was higher in Cohort 3 than in Cohort 1 (Figure 1A; Table S9). Within‐cohort increases at 12 months were observed in Cohorts 1 and 2 but not in Cohort 3. The proportion with HbA1c >7% is shown in Figure 1B.

Older age was independently associated with lower HbA1c, whereas female sex was associated with higher HbA1c.

Glucose‐lowering therapies are summarised in Table 2. Insulin and dipeptidyl peptidase‐4 (DPP‐4) inhibitors were the most frequently prescribed agents. Metformin use was more common in Cohort 3. GLP‐1 receptor agonists were introduced earlier and more frequently in Cohort 1, whereas use in Cohort 3 during early follow‐up was limited. SGLT2 inhibitor use was low across cohorts. Prednisone exposure was universal in Cohort 3 and less frequent in Cohorts 1 and 2.

TABLE 2.

Antihyperglycaemic and antihypertensive medication use during the first year after kidney transplantation, by cohort.

	Cohort 1 (n = 99)				Cohort 2 (n = 81)				Cohort 3 (n = 40)
	B	3M	6M	12M	B	3M	6M	12M	B	3M	6M	12M
Antihyperglycaemic medications, %
Insulin	90.9*	92.6*	91.4	89.5	82.7*	80.2*	82.5	83.8	65.0*	77.5*	82.1	86.5
Metformin	0.0	3.2	3.4*	2.6*	0.0	1.2	2.5*	5.4*	0.0	5.0	15.4*	16.2*
SFU	5.1	16.0	13.8	10.5	4.9	8.6	10.0	10.8	12.5	5.0	5.1	10.8
DPP‐4i	16.2*	29.8	34.5	42.1	23.5*	25.9	22.5	21.6	57.5*	25.0	30.8	24.3
SGLT2i	0.0	2.1	3.4	5.3	0.0	0.0	2.5	5.4	0.0	2.5	0.0	8.1
GLP‐1 RA	10.1*	19.1*	22.4*	26.3	6.2*	12.3*	15.0*	13.5	0.0*	0.0*	0.0*	8.1
Antihypertensive medications, %
CCB	53.5	47.9	48.3	52.6	51.9	55.6	55.0	51.3	57.5	62.5	56.4	67.6
Diuretics	22.2*	12.8*	10.3*	15.8*	30.9	22.2*	22.5*	24.3*	67.5*	32.5*	38.5*	37.8*
ARB/ACEI	1.0	3.2	10.3	7.9*	0.0	3.7	7.5	5.4*	42.5*	5.0	12.8	21.6*
BB	62.6	64.9	67.2	63.2	55.6	58.0	60.0	59.5	60.0	62.5	64.1	59.5
Hydralazine	16.2*	5.3*	17.2*	15.8*	13.6*	16.0*	7.5*	5.4	5.0*	2.5*	2.6*	2.7*
Clonidine/Doxazosin	5.1*	7.4*	6.9*	2.6*	7.4*	7.4*	2.5*	2.7*	30.0*	22.5*	30.8*	48.6*

Note: Values represent the percentage of participants receiving each medication at each time point.

Abbreviations: 3M, 3 months; 6M, 6 months; 12M, 12 months; ACEI, angiotensin‐converting enzyme inhibitors; ARB, angiotensin II receptor blockers; B, first post‐transplant visit; BB, beta‐blockers; CCB, calcium channel blockers; DPP‐4i, dipeptidyl peptidase‐4 inhibitors; GLP‐1 RA, glucagon‐like peptide‐1 receptor agonists; SFU, sulfonylureas; SGLT2i, sodium‐glucose co‐transporter‐2 inhibitors.

^* p < 0.01 for comparisons of Cohort 1 or 2 versus Cohort 3 at the same time point.

3.3.2. Body mass index

BMI remained stable over time with no between‐cohort differences (Figure 2A; Table S9). Older age and longer diabetes duration were associated with lower BMI. The prevalence of BMI ≥30 kg/m² remained ≥30% throughout follow‐up (Figure 2B).

FIGURE 2.

Longitudinal changes in body mass index (BMI) after kidney transplantation. (A) Adjusted mean BMI (95% CI) at baseline, 3, 6, and 12 months by cohort estimated from linear mixed‐effects models. No significant changes over time or between cohorts were observed. (B) Proportion of participants with BMI ≥30 kg/m² at each visit by cohort.

3.3.3. Blood pressure control

Neither SBP nor DBP changed significantly over time or differed between cohorts. Longer dialysis duration was associated with lower SBP and DBP, whereas steroid exposure was associated with higher SBP. Older age and female sex were associated with lower DBP (Tables S3 and S4).

Consistent with these findings, Tukey‐adjusted within‐and between‐cohort comparisons showed no significant differences at any visit (Tables 3 and S8).

TABLE 3.

Adjusted marginal means (95% CI) for blood pressure and lipid parameters during the first year after kidney transplantation, by cohort.

Cohort	Baseline	3 months	6 months	12 months
SBP (mmHg)
1	139.9 (135.8–144.1)	141 (136.9–145.2)	138.6 (134.5–142.7)	138.6 (134.4–142.7)
2	142.5 (137.9–147.2)	142.6 (138–147.2)	140.3 (135.7–144.9)	138.5 (133.9–143.1)
3	137.5 (130.2–144.8)	140.7 (133.4–148)	137.2 (129.9–144.5)	143.2 (135.9–150.5)
DBP (mmHg)
1	68.1 (65.4–70.7)	69.2 (66.6–71.9)	69 (66.3–71.6)	70.3 (67.7–73)
2	70.4 (67.4–73.3)	71.6 (68.7–74.6)	71.6 (68.6–74.6)	69.4 (66.5–72.4)
3	68.8 (64–73.5)	69.2 (64.5–74)	67.9 (63.1–72.7)	70.6 (65.8–75.3)
Total cholesterol
1	146.7 (136.4–157)	140.8 (130.6–151.1)	159.2 (148.9–169.5)	164.7 (154.5–174.9)
2	147.8 (136.3–159.2)	147.1 (135.6–158.6)	172 (160.5–183.4)	166.3 (154.8–177.8)
3	181.8 (163.5–200.1)**	151.6 (133.3–169.9)*	149.2 (130.9–167.6)*	148.1 (129.7–166.4)*
LDL‐C (mg/dL)
1	71 (62.6–79.3)	67.2 (58.8–75.5)	81.7 (73.4–90.1)	74.8 (66.4–83.1)
2	68.6 (59.4–78.1)	73.5 (64.2–82.8)	86.8 (77.5–96.2)	86.5 (77.1–95.8)
3	87.9 (72.9–102.9)	72.9 (57.9–87.9)	72.3 (57.3–87.3)	69 (54–84)
HDL‐C (mg/dL)
1	45.2 (40.8–49.5)	51.6 (47.4–55.7)	49.9 (45.7–54.1)	58.1 (53.9–62.3)*
2	45.1 (40.4–49.7)	48.9 (44.3–53.6)	50.9 (46.2–55.5)	54.1 (49.4–58.7)
3	41.9 (34.4–49.3)	44.2 (36.7–51.6)	46.9 (39.4–54.3)	48.3 (40.9–55.8)
TG (mg/dL)
1	173.4 (150.7–196)	132.6 (109.9–155.2)*	152.6 (130–175.3)	175.7 (153–198.3)
2	179 (153.6–204.4)	151.4 (126–176.7)	162.9 (137.6–188.3)	188 (162.7–213.4)
3	213.9 (172.7–255)**	165 (123.9–206.1)	169.2 (128.1–210.4)	190.9 (149.8–232)

Note: Values are adjusted marginal means (95% CI) from linear mixed‐effects model with a random intercept for participant, adjusted for age, sex, diabetes duration, dialysis duration, transplant hospitalisation length of stay, and steroid exposure.

Abbreviations: DBP, diastolic blood pressure; HDL‐C, high‐density lipoprotein cholesterol; LDL‐C, low‐density lipoprotein cholesterol; SBP, systolic blood pressure; TG, triglycerides.

^* p < 0.05 versus baseline within the same cohort.

^** p < 0.05 versus Cohort 1 at the same time point.

At 12 months, SBP >130 mmHg affected approximately 50% of recipients in Cohorts 1–2 and >75% in Cohort 3. DBP >80 mmHg was present in 33%–40%. Antihypertensive medication use is summarised in Table 2.

3.3.4. Lipid profile

Total cholesterol increased at 6 and 12 months (Table S5). Baseline values were higher in Cohort 3 but declined over time. Older age and longer diabetes duration were associated with lower total cholesterol, and female sex was associated with higher levels. HDL‐C increased significantly at all follow‐up visits. Steroid exposure was independently associated with higher HDL‐C. At 12 months, LDL‐C ≥70 mg/dL occurred in 21%, 63%, and 42% in Cohorts 1–3.

Triglyceride declined at 3 months only. Statin therapy was common across cohorts.

4. DISCUSSION

In this multicentre cohort of kidney transplant recipients with pre‐existing diabetes, metabolic trajectories during the first post‐transplant year were characterised primarily by stabilisation rather than consistent improvement. Across three care models, including an endocrinology‐integrated transplant clinic, glycaemic control, body mass index, blood pressure, and atherogenic lipid parameters frequently remained above recommended targets. Differences between care models were modest.

4.1. Metabolic trajectories

Glycaemic control did not improve over time and worsened modestly by 12 months, with more than half of the recipients exceeding HbA1c targets at 1 year. Blood pressure trajectories were similarly stable, with over half remaining above systolic targets. BMI showed little change, yet obesity (BMI ≥30 kg/m²) persisted in approximately one‐third of participants.

Lipid changes were heterogeneous. HDL‐C increased over time, whereas LDL‐C showed a transient mid‐year rise that did not persist, and triglyceride reductions were not sustained. Taken together, these findings suggest partial metabolic recovery after transplantation but limited overall cardiometabolic optimisation.

4.2. Glycaemic management

The absence of sustained glycaemic improvement likely reflects the combined effects of immunosuppressive therapy, post‐transplant weight changes, and progressive β‐cell dysfunction.^{10, 11} Although HbA1c has limitations early after transplantation, ¹² it remains the most practical longitudinal metric for routine care.

Earlier uptake of GLP‐1 receptor agonists in the ETC cohort indicates that endocrinology involvement may facilitate access to contemporary therapies. However, overall use of GLP‐1 RAs and SGLT2 inhibitors remained limited. Given emerging evidence supporting their safety and cardiometabolic benefits in transplant recipients,^{13, 14, 15, 16} cautious expansion of real‐world use represents a key opportunity for future optimisation.

4.3. Weight, blood pressure, and lipid management

Persistent obesity and hypertension remain clinically relevant, as both are major contributors to cardiovascular disease, a leading cause of death among diabetic kidney transplant recipients.^{17, 18, 19, 20} In this study, 27%–40% of patients had coronary artery disease and 15%–29% had heart failure at transplantation. Obesity affected more than 30% of recipients across cohorts, and excess weight gain is associated with increased graft failure and mortality risk.^{21, 22}

Blood pressure control did not improve over time. More than half recipients remained hypertensive at all visits, reflecting a sustained atherosclerotic burden. ²³ Although optimal post‐transplant blood pressure targets remain debated, effective blood pressure control is strongly associated with graft loss and cardiovascular mortality.^{12, 24, 25, 26} These findings suggest that current strategies may be insufficient to meaningfully alter weight and blood pressure trajectories in this population.

Despite high statin use, LDL‐C did not demonstrate sustained improvement, while HDL‐C increases may reflect metabolic recovery or steroid exposure rather than care model differences. Overall, lipid patterns indicate partial but incomplete risk modification. PCSK9 inhibitors, which appear safe and effective in transplant recipients, ²⁷ were not used and warrant consideration.

4.4. Care model comparison

Metabolic trajectories were broadly similar across cohorts. While the endocrinology‐integrated model supported structured follow‐up and earlier use of newer therapies, large differences in metabolic outcomes were not observed. in a population with complex comorbidities and chronic immunosuppression, preventing deterioration may itself be clinically meaningful; however, these findings suggest that specialist involvement alone may be insufficient without protocol‐driven intensification and structured lifestyle interventions.

4.5. Healthcare context

Baseline microvascular and macrovascular complication burden was similar across cohorts, suggesting that geographic location and healthcare system had minimal influence on diabetes severity among transplant candidates in developed countries.

Consistent with prior reports, men received kidney transplants at approximately twice the rate of women,^{28, 29, 30, 31, 32} reflecting persistent disparities in transplant access.^{33, 34, 35} Time on dialysis before transplantation was significantly longer in the United States, a factor associated with poorer long‐term outcomes^{36, 37, 38} whereas post‐transplant hospital stays were longer in Spain. These differences likely reflect structural contrasts between healthcare systems rather than differences in care quality. Importantly, these system‐level variations did not translate into clearly divergent metabolic trajectories.

4.6. Limitations

Missing data increased over time, although multiple imputation and sensitivity analyses supported robustness. Cohort sizes were unequal, and residual confounding cannot be excluded despite multivariable adjustment. The observational design precludes causal inference, and findings may not be generalisable to centres with different transplant practices or patient populations.

5. CONCLUSIONS

During the first year after kidney transplantation, metabolic control was characterised by stabilisation rather than consistent improvement, and cardiometabolic risk factors frequently remained above recommended targets despite specialised care. These findings underscore the persistent metabolic vulnerability of kidney transplant recipients with diabetes and highlight the need for earlier, protocolised, and multidisciplinary metabolic management strategies. Prospective studies are needed to determine whether structured lifestyle interventions and broader use of contemporary pharmacotherapies can meaningfully improve cardiovascular and graft outcomes in this high‐risk population.

FUNDING INFORMATION

Antia Fernandez‐Pombo was funded by a Juan Rodes research contract (JR23/00042) from the Instituto de Salud Carlos III and co‐financed by the European Regional Development Fund.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

Supporting information

Data S1. Supporting Information.

DATA AVAILABILITY STATEMENT

The data presented in the study are available on request from the corresponding authors. These data are not publicly available due to privacy concerns relating to personal clinical information.