https://orcid.org/0000-0001-6143-461X
Sodium–glucose co-transporter 2 inhibitors (SGLT2is) have been revolutionary in the care of patients with chronic kidney disease (CKD). Initially developed for diabetes control, these drugs have been shown to be effective at preventing both diabetic and non-diabetic CKD progression in several large randomized controlled trials (RCTs). Consequently, they are now considered first-line treatment in CKD. The underlying mechanism of their protective effect is not fully understood, but in addition to haemodynamic factors, important metabolic and anti-inflammatory effects are likely to play a role (Fig. 1). A protective effect of dapagliflozin has been observed in preclinical models of acute kidney injury (AKI) induced by ischaemia–reperfusion, radiocontrast and lipopolysaccharides (LPS) in both diabetic and non-diabetic mice [1]. Benefits offered by dapagliflozin appear mainly mediated by the promotion of kidney protective pathways, including antioxidant defences, mitochondrial biogenesis and preservation of Klotho expression [1]. In RCTs and real-life studies, gliflozins are associated with a lower risk of AKI [2], different from renin–angiotensin–aldosterone system (RAAS) blockers, another class of nephroprotective drugs with haemodynamic effects.
Presumed renal effects of SGLT2i therapy.
SGLT2 inhibition appears to be protective after AKI as well. In a retrospective cohort of 21 330 eligible US veterans with type 2 diabetes mellitus and proteinuria hospitalized with AKI, 6562 patients received a SGLT2i (31%) over a median follow-up of 2.1 years after hospitalisation. Patients treated with an SGLT2i had a lower mortality independent of kidney function recovery [3].
Among 230 366 patients with acute kidney disease (AKD), those who used SGLT2i had a significantly lower risk of mortality, major adverse kidney events and major adverse cardiovascular events [4]. Finally, a single-centre retrospective analysis of 265 patients receiving contrast showed that patients on SGLT2i therapy had a lower risk of AKI and lower need for kidney replacement therapy (KRT) [5].
These results raise the question whether there is a role for SGLT2i to prevent complications and reduce mortality in acutely ill patients. This question was addressed in the DEFENDER trial (NCT05558098), a multicentre, unblinded, placebo-controlled RCT in 22 intensive care units (ICUs) across Brazil [6]. Critically ill patients not on chronic SGLT2i treatment with at least one organ dysfunction and a projected stay of >48 h were included. A total of 507 participants were randomized to receiving dapagliflozin 10 mg versus placebo for 14 days after randomization. The primary outcome was a hierarchical composite of hospital mortality, initiation of KRT and ICU length of stay through 28 days after randomization.
There was no difference in the primary endpoint: dapagliflozin did not increase the win ratio for the hierarchical endpoints of hospital mortality and ICU length of stay. Among all secondary endpoints, the highest probability of benefit was seen regarding the use of KRT. A total of 10.9% of patients in the dapagliflozin group received KRT versus 15.1% of the placebo group. However, the adjusted odds ratio (OR) for the Bayesian model did not reach statistical significance [OR 0.76 (95% confidence interval 0.50–1.18)]. Importantly, there was no difference in side effects, and no cases of ketoacidosis were reported. Patients were not fasting during the trial.
WHAT DOES THE DEFENDER TRIAL TELL US?
The DEFENDER trial tells us, first, that administering dapagliflozin to critically ill patients with organ dysfunction does not reduce mortality or length of stay in hospital. Second, SGLT2 inhibition in severely acutely ill patients is safe, with no increased risk of complications or episodes of euglycaemic ketoacidosis.
Although the trial was negative, it broadens our knowledge of SGLT2 inhibition in acutely ill patients, reinforcing that the treatment is generally safe in this population. It further suggests that dapagliflozin use may be associated with lower KRT administration, but more data are necessary.
Of note, in the RECOVERY study (NCT04381936), empagliflozin use in COVID-19 patients was generally safe but also did not result in lower mortality, and two cases of ketoacidosis were observed [7]. A recent cohort study even demonstrated that maintaining SGLT2i treatment in hospitalized diabetic patients was associated with lower mortality and shorter length of stay [8]. This suggests that the practice of discontinuing SGLT2i at admission to hospital may not be appropriate and that it may be safe to continue them throughout hospitalization and potentially during severe acute illness.
CAN SGLT2 INHIBITION PREVENT AKI OR KRT DURING HOSPITALIZATION?
This is still an open question that will require well-designed prospective trials with a focus on AKI risk stratification, possibly using specific enrichment strategies to avoid patient heterogeneity at enrolment.
CAN SGLT2is PREVENT AKI-TO-CKD TRANSITION?
Preclinical data [9] and observational studies support this hypothesis, however, the protective effect of SGLT2i on AKI-to-CKD progression needs to be investigated in future RCTs with specific emphasis on the post-AKI period. Of note, recent studies have suggested a protective effect of SGLT2i on cellular senescence [10], a state of growth arrest characterized by the release of inflammatory mediators that plays a key role in the loss of mature tubular epithelial cell phenotypes and the development of early fibrosis during AKI-to-CKD transition.
SHOULD WE CHANGE OUR GENERAL PRACTICE?
Given the DEFENDER trial was negative, using gliflozin in all ICU patients cannot be recommended. Obviously it needs to be acknowledged that the patients who participated in the DEFENDER trial were critically ill with at least one organ dysfunction and an expected stay in the ICU of ≥48 h. In this cohort, AKI is a frequent complication, affecting >50% of patients. The pathophysiology is heterogeneous and complex, including haemodynamic instability, microcirculatory and endothelial dysfunction, tubular injury, inflammation, complement activation and altered mitochondrial function. It is possible that any potential benefits of SGLT2 inhibitors were confounded by this heterogeneity and high severity of illness.
The authors acknowledged some other limitations that may have also impacted the results, such as the unblinded nature of the trial and the inclusion of an unselected population of critically ill patients with various stages of acute illness. Further, dapagliflozin was given either orally or enterally, which may have affected systemic concentrations and efficacy.
Despite the negative results, the trial sets the stage for research in patients at high AKI risk or with early AKI to prevent KRT or to mitigate the AKI-to-CKD transition. Further, the frequent practice of stopping SGLT2i in acutely ill or hospitalized patients should be questioned. Patients on chronic SGLT2i therapy were excluded from the DEFENDER trial. Observations in the diabetic population suggest that stopping these drugs in high-risk patients with a clear indication is probably unnecessary and potentially harmful. With regards to the question whether SGLT2is prevent AKI/KRT, more research is needed. It is also not possible to say whether continuation of chronic SGLT2i therapy has different effects on kidney function than new administration.
In summary, the DEFENDER trial broadens our understanding of SGLT2 inhibition in acutely ill patients and suggests that these drugs are safe but not beneficial in such patients. We now hope to see more clinical trials testing these drugs for AKI prevention or treatment and for mitigating the AKI-to-CKD transition.
ACKNOWLEDGEMENTS
The ERAKI Working Group is an official body of the European Renal Association.
FUNDING
None.
CONFLICT OF INTEREST STATEMENT
None declared.
REFERENCES
Author notes
These authors contributed equally.
Comments