Sodium glucose cotransporter 2 inhibitors are associated with renal stabilization in heart transplantation

Abstract

Sodium glucose cotransporter 2 inhibitors (SGLT2i) are standard of care for type 2 diabetes mellitus, heart failure, and chronic kidney disease (CKD). Heart transplant (HTx) recipients are at increased risk of diabetes and CKD, and both are independently associated with increased mortality. In a retrospective analysis of 104 HTx recipients with diabetes (23 exposed to SGLT2i, 81 not exposed), SGLT2i treatment was associated with stable renal function at 3 years post-HTx, measured by estimated glomerular filtration rate change from baseline (median change of 0 ml/min/1.73 m² (interquartile range [IQR] −13 to +11)), compared to a change of −15 ml/min/1.73 m² (IQR −27 to +1) in patients not exposed to SGLT2i (p = 0.02). There was no significant difference in survival by SGLT2i exposure, adjusted for diabetes type and baseline creatinine (hazard ratio 0.34, confidence intervals 0.11-1.06, p = 0.06). Further investigation of SGLT2i in HTx recipients, particularly focusing on renal outcomes, is required.

Heart transplant (HTx) recipients are at increased risk of chronic kidney disease (CKD) and progressive renal impairment, particularly in the setting of comorbid diabetes.1, 2 When present, both renal function decline and diabetes are independently associated with adverse transplant outcomes, including premature mortality.1, 2, 3 Sodium glucose cotransporter 2 inhibitors (SGLT2i) are approved to treat type 2 diabetes mellitus (T2DM) and CKD in the nontransplant population, and we have previously reported glycemic benefits associated with SGLT2i treatment of diabetes in HTx recipients.4, 5 However, there are no studies examining renal outcomes in HTx patients exposed to SGLT2i treatment. This study assessed renal outcomes in HTx recipients with diabetes exposed to SGLT2i (commenced within 12 months of HTx) compared to HTx recipients with diabetes not exposed to SGLT2i treatment. The association of early SGLT2i exposure with survival was also assessed.

We conducted a retrospective observational study of patients who underwent HTx between January 2015 and December 2020 at a single center. Data were extracted from hospital electronic medical records, with follow-up until December 2023. Ethics approval was provided by the Human Research and Ethics Committee, St Vincent’s Hospital Sydney (2022/ETH01323).

All HTx recipients in the study period with T2DM or post-transplant diabetes were included. To be included in the SGLT2i-exposed cohort, patients must have commenced SGLT2i within 12 months of HTx and received SGLT2i treatment for a minimum of 6 months. Patients who were dialysis dependent from the time of transplant or who received a combined heart-kidney transplant were excluded. Renal function was assessed by estimated glomerular filtration rate (eGFR), calculated using the Chronic Kidney Disease Epidemiology Collaboration equation, and recorded at the time of HTx and annually thereafter. The primary outcome was the between-groups difference in eGFR change from baseline assessed at 3 years post-HTx.

Data are expressed as median and interquartile range (IQR) or number and percentage. Nonparametric comparisons were made using Mann-Whitney test and binary outcomes using Fisher’s exact test. Associations between SGLT2i exposure and survival were assessed by Cox proportional hazards, adjusted for demographic factors that were statistically different at baseline (p < 0.05). Results of Cox proportional regression are presented as hazard ratio and associated 95% confidence intervals. Results were censored at the time of death, participants who were deceased at 3 years were not included in the primary renal outcome. Analyses were conducted using SPSS version 28 (SPSS Inc., Chicago, IL), and figures were created using GraphPad Prism version 10 (GraphPad Software, Boston, MA).

Local immunosuppression protocol included tacrolimus with trough level target of 6 to 10 mg/liter, and mycophenolate mofetil 1,000 mg twice daily. Prednisolone was initiated at 1 mg/kg/day and decreased to 0.3 mg/day by day 21; thereafter, corticosteroid dose was at the discretion of the treating physician. The introduction of everolimus was at clinician discretion. Acute rejection was treated with pulse corticosteroids. Endocrinology review was available for all patients through a clinic linked to the transplant service.

During the study period, 104 patients met the inclusion criteria. Of 104 HTx recipients, 23 (22%) were exposed to SGLT2i and 81 (78%) were not exposed to SGLT2i. In the exposed group, SGLT2i treatment was commenced within 6 months of HTx in 13 (57%) patients, including 10 (48%) individuals in whom SGLT2i treatment was initiated in the first 3 months; median time to commencement was 6 months (IQR 1-9). Median duration of SGLT2i exposure was 44 months (IQR 17-63). SGLT2i treatment was ceased in 9 (39%) patients after a median of 16 months (IQR 11-31). The reasons for cessation were reduction in eGFR <30 ml/min/1.73 m² as per guidelines at the time (n = 6), respiratory sepsis (n = 1), weight loss (n = 1), and improved glycemia (n = 1).

At baseline, cohorts were similar in age, sex, and underlying cause of heart failure (Table 1). Patients exposed to SGLT2i were more likely to have pre-existing T2DM (n = 16, 70%) compared to nonexposed, who were more likely to have post-transplant diabetes (n = 58, 72%) (p < 0.001). The SGLT2i-exposed cohort was more likely to have also been exposed to SGLT2i before transplant than the nonexposed cohort (43% vs 10%, p < 0.001). The SGLT2i-exposed cohort had higher median creatinine 115 μmol/liter (IQR 93-138) versus 95 μmol/liter (IQR 77-119) in those not exposed to SGLT2i (p = 0.03). Post-transplant, there were similar rates of everolimus use and episodes of ≥2R rejection in the 2 cohorts.

Table 1.

Characteristics and Outcomes

Characteristics	SGLT2i commenced within first 12 months (n = 23)	No SGLT2i (n = 81)	p-value
Baseline characteristics
Age (years); median (IQR)	58 (48-63)	54 (47-62)	0.33
Male; n (%)	20 (87)	56 (69)	0.11
eGFR (ml/min/1.73 m2)	61 (50-71)	73 (54-99)	0.053
Creatinine (μmol/liter)	115 (93-138)	95 (77-119)	0.03
Underlying cardiomyopathy; n (%)			0.83
Ischemic	9 (39)	20 (25)
Dilated	8 (35)	29 (36)
Familial	2 (9)	9 (11)
Infiltrative	0 (0)	4 (5)
Toxic	0 (0)	6 (7)
Congenital	1 (4)	4 (5)
Viral	2 (9)	5 (6)
Other	1 (4)	4 (5)
Diabetes type; n (%)			<0.001
Type 2 diabetes	16 (70)	23 (28)
Post-transplant diabetes	7 (30)	58 (72)
Pretransplant SGLT2i exposure	10 (43)	8 (10)	<0.001
Post-transplant characteristics
Everolimus exposure	16 (70)	52 (64)	0.80
Time to commencement of everolimus (months)	8 (5-15)	8 (5-16)	0.90
Rejection (≥2R)	8 (35)	28 (35)	0.99

Abbreviations: eGFR, estimated glomerular filtration rate; IQR, interquartile range; SGLT2i, sodium glucose cotransporter 2 inhibitors.

Over 3 years of follow-up post-HTx, SGLT2i treatment was associated with a median eGFR change of 0 ml/min/1.73 m² (IQR −13 to +11) from baseline, compared to −15 ml/min/1.73 m² (IQR −27 to +1) in patients not exposed to SGLT2i (p = 0.02) (Figure 1). The 3-year mortality rate was 39% (n = 9) in the SGLT2i cohort and 30% (n = 24) in the nonexposed cohort (p = 0.45). There was no difference in survival by SGLT2i exposure, adjusted for diabetes type and baseline creatinine (hazard ratio 0.34, confidence intervals 0.11-1.06, p = 0.06) (Figure 2). There were no serious adverse events related to SGLT2i treatment. No cases of diabetic ketoacidosis or Fournier’s gangrene occurred in SGLT2i-exposed patients, although urinary tract infection occurred in 1 patient.

Figure 1.

(A) The between-groups difference in eGFR change from baseline assessed at 3 years post heart transplantation, comparisons made using Mann-Whitney test. (B) Median and interquartile range of change in eGFR from baseline over 3 years of follow-up after heart transplantation. eGFR, estimated glomerular filtration rate; SGLT2i, sodium glucose cotransporter 2 inhibitors.

Figure 2.

Adjusted survival (months) in heart transplant recipients with diabetes exposed to SGLT2i versus those not exposed, adjusted for diabetes type and baseline creatinine. CI, confidence intervals; HR, hazard ratio; SGLT2i, sodium glucose cotransporter 2 inhibitors.

Over 3 years of follow-up, initiation of SGLT2i within 12 months of HTx was associated with stable renal function in HTx recipients with diabetes, compared to a significant decline in patients not exposed to SGLT2i. Separation of eGFR curves occurred early, within the first 12 months of transplant. Overall survival did not differ between groups, although there was a trend toward increased survival associated with SGLT2i exposure that may have manifested with longer follow-up or a larger study cohort. Overall, our hypothesis-generating retrospective study reports the first data of long-term renal outcomes associated with SGLT2i treatment in HTx recipients with diabetes and is concordant with SGLT2i outcome data observed in nontransplant patient settings.⁶

Kidney dysfunction is a frequent complication of HTx, with most recipients developing stage 3 or greater CKD during long-term follow-up.1, 7, 8 Very few treatments for CKD progression post-HTx have been identified apart from retrospective data supporting optimization of hypertension and eventual kidney transplantation.⁹ Impaired renal function at the time of transplant and T2DM are associated with higher rates of CKD progression in HTx recipients.1, 7, 8 In our study, despite high rates of pretransplant T2DM and CKD, SGLT2i-exposed patients had no appreciable loss of renal function, suggesting SGLT2i treatment may mitigate the negative impact of transplantation on renal function in patients with these risk factors. The high rate of SGLT2i exposure before transplant may represent a legacy effect and/or pretransplant optimization, with previous data suggesting a survival association with SGLT2i use before HTx.¹⁰

Limitations of our study are inherent to the study design and small sample size. Based on our data, causality cannot be established, and although benefit may be a direct effect of SGLT2i treatment, results may also be related to other factors such as diabetes optimization, other agents, or an unmeasured selection bias. Decision to commence SGLT2i treatment was not randomized and was made at discretion of the treating endocrinologist or cardiologist. Factors contributing to that decision, such as infection risk and peripheral vascular disease, are not elucidated in this study. The small patient population limited power and further statistical analysis. The study was not powered to examine differences in survival. Baseline renal function was based on pretransplant results to allow standardization in this cohort; however, further research could use a post-transplant renal function which may represent a new baseline.

In conclusion, we found that SGLT2i exposure for at least 6 months commenced within a year of HTx was associated with amelioration in the decline in renal function typically observed in HTx recipients with diabetes. Further investigation of this important class of medications is necessary, and prospective randomized trials of SGLT2i in HTx recipients are underway.¹¹ HTx recipients have been excluded from major cardiovascular and renal trials of SGLT2i, yet with increased rates of diabetes and CKD post-transplant, HTx recipients are a prime patient cohort and likely have much to gain from SGLT2i treatment.

CRediT authorship contribution statement

Lisa M. Raven: Conceptualization, Methodology, Validation, Formal analysis, Investigation, Writing - original draft, Visualization. Jerry R. Greenfield: Conceptualization, Methodology, Visualization, Supervision, Writing – review & editing. Andrew Jabbour: Writing – review & editing. Peter S. Macdonald: Writing – review & editing. Christopher A. Muir: Conceptualization, Methodology, Visualization, Supervision, Writing – review & editing.

Disclosure statement

Lisa M. Raven reports financial support was provided by National Heart Foundation of Australia (Award Reference 107026). Lisa M. Raven reports a relationship with AstraZeneca that includes travel reimbursement. Jerry R. Greenfield reports a relationship with Novo Nordisk that includes speaking and lecture fees and provision of semaglutide and placebo for a study in type 1 diabetes. Peter S. Macdonald reports a relationship with AstraZeneca that includes consulting or advisory. Peter S. Macdonald reports a relationship with Boehringer Ingelheim Ltd. that includes consulting or advisory. Peter S. Macdonald reports a relationship with Novartis that includes consulting or advisory. Peter S. Macdonald reports a relationship with Eli Lilly that includes consulting or advisory. Christopher A. Muir reports a relationship with Boehringer Ingelheim Ltd. that includes speaking and lecture fees and travel reimbursement. Christopher A. Muir reports a relationship with Eli Lilly that includes speaking and lecture fees and travel reimbursement. Christopher A. Muir reports a relationship with Novo Nordisk that includes speaking and lecture fees and travel reimbursement. The other authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. There was no involvement from funding bodies in the study design, conduct, and reporting of this work.