Among patients with type 2 diabetes (T2D) recovering from severe acute kidney injury (AKI) requiring dialysis, initiation of an SGLT2 inhibitor within 30 days after discharge was associated with better kidney outcomes than initiation 31-90 days after discharge, new research showed.

The study “extends prior evidence by suggesting that timing during post-AKI recovery may matter in a population that is largely underrepresented in previous trials,” first author Yu-Ting Chou, MD, of the Division of Nephrology, Department of Internal Medicine, National Taiwan University Hospital in Taipei, Taiwan, told Medscape Medical News.

Importantly, “acute kidney disease may represent a pivotal therapeutic window between acute kidney injury and subsequent CKD [chronic kidney disease] progression, and our findings suggest that earlier SGLT2 inhibitor initiation during this period may help preserve kidney function,” Chou added.

The study was published in JAMA Network Open.

Compounding Risks After Severe AKI

AKI increases the risk for subsequent kidney disease progression and cardiovascular events, with risks further elevated when AKI is severe enough to require dialysis and in patients with T2D, the authors explained.

Although SGLT2 inhibitors have shown benefits in patients with T2D and CKD, patients recovering from severe AKI have often been excluded from major randomized trials, leaving questions about the efficacy and optimal timing of treatment in this population.

Treatment timing is important because clinicians may delay SGLT2 inhibitor treatment following AKI due to concerns about diabetic ketoacidosis, volume depletion, infection, or hemodynamic instability, Chou explained.

Conversely, delaying treatment could hinder the potential kidney and cardiovascular benefits of SGLT2 inhibitors in patients who are clinically stable after discharge.

“Evidence addressing this clinical decision window remains limited,” the authors wrote.

Comparing SGLT2 Inhibitor Outcomes

To investigate, the authors conducted a cohort study using a target trial emulation design. They identified 129,179 eligible patients with T2D in the TriNetX Global Collaboration Network who had been hospitalized for AKI requiring dialysis and had discontinued dialysis after discharge between January 2015 and September 2024. None had a record of receiving an SGLT2 inhibitor prescription in the prior year.

Of these patients, only 4406 (3.4%) initiated an SGLT2 inhibitor within 90 days of discharge, among whom 51.5% started treatment within 30 days and 48.5% started between days 31 and 90.

After 1:1 propensity score matching, the analysis included two groups of 1912 patients each: those who initiated SGLT2 inhibitors within 30 days and those who initiated treatment 31-90 days after discharge.

Patients in the two groups had a mean age of approximately 63 years, and roughly 58% were men.

After a mean follow-up of 3.4 years, patients who initiated SGLT2 inhibitors within 30 days had a significantly lower risk for adverse kidney events, defined as dialysis reinitiation, incident end-stage kidney disease, or a recorded estimated glomerular filtration rate < 15 mL/min/1.73 m² compared with those who initiated treatment 31-90 days after discharge (3.9% vs 6.2%; adjusted hazard ratio [aHR], 0.62; P = .001).

There were no significant between-group differences in the other co-primary endpoints of all-cause mortality (aHR, 1.06; P = .63) or major adverse cardiovascular events (aHR, 1.03; P = .81), nor were there differences in selected 6-month safety outcomes.

Of note, the study did not assess the duration of SGLT2 inhibitor therapy.

GLP-1 Added Benefit?

The researchers also conducted an exploratory subgroup analysis and found a stronger association between early SGLT2 inhibitor initiation and improved kidney outcomes among those who also received GLP-1 receptor agonists (aHR, 0.13) compared with those not receiving GLP-1 receptor agonists (aHR, 0.84; P = .004 for interaction).

GLP-1 receptor agonist use could have preceded or overlapped with SGLT2 inhibitor initiation and therefore did not necessarily represent simultaneous initiation of combination therapy, Chou noted.

“Because these subgroup analyses were exploratory, these findings should be interpreted cautiously and viewed as hypothesis-generating,” the authors wrote.

“Nevertheless, the association observed among GLP-1 receptor agonist users suggests a potential synergistic kidney-protective effect when early SGLT2 inhibitor initiation is combined with GLP-1 receptor agonist therapy.”

Potential Mechanisms

Key mechanisms proposed to underlie the benefits of SGLT2 inhibitors in patients with stable CKD and AKI included reduced intraglomerular pressure; improved renal oxygen use; and the “attenuation of inflammatory, oxidative, endothelial, and metabolic stress pathways,” the authors wrote. “These mechanisms may be relevant during recovery from severe AKI, in which maladaptive repair and progressive nephron loss may contribute to transition from AKI to CKD.”

However, patients with AKI requiring dialysis could have widely varying recovery trajectories, which may help explain the low rate (3.4%) of SGLT2 inhibitor initiation in this population.

The low rate “is consistent with clinical practice, where many clinicians remain conservative about initiating SGLT2 inhibitors soon after AKI recovery,” Chou said.

Previous research comparing SGLT2 inhibitor use with nonuse among patients with T2D recovering from dialysis-requiring AKI had shown lower risks for mortality, major adverse cardiovascular events, and other cardiovascular events with SGLT2 inhibitor use, but randomized trials were lacking, Chou noted.

The new findings add to these data, raising “the possibility that the early post-AKI recovery period may be a clinically relevant time frame for SGLT2 inhibitor initiation and support prospective randomized clinical trials to clarify the optimal timing, efficacy, and safety of treatment,” the authors concluded.

Notable Caveats

Commenting to Medscape Medical News, Ashish Upadhyay, MD, clinical professor of medicine at Boston Medical Center, Boston University Chobanian & Avedisian School of Medicine in Boston, called the study “important, clinically interesting, and hypothesis-generating.”

“It supports the idea that clinicians should not necessarily delay SGLT2 inhibitor initiation simply because a patient recently recovered from dialysis-requiring AKI,” he said.

“However, I would not interpret this study as proving that initiation within 30 days is superior or that 30 days represents a biologically meaningful threshold,” Upadhyay said.

One key limitation was the potential for confounding by “clinical readiness,” he explained.

“Patients who were started within 30 days after discharge may differ in ways that are difficult to capture in a database,” Upadhyay said.

Although the study used appropriate matching, physicians may have selected relatively healthier patients for earlier SGLT2 inhibitor initiation, such as those with more complete or rapid kidney recovery or hemodynamic stability, he explained.

“While well done, observational studies have limitations, and I would consider this study as providing a strong rationale for a prospective randomized trial examining when SGLT2 inhibitors should be initiated after AKI recovery,” Upadhyay concluded.

The authors reported no relevant disclosures. Upadhyay reported consulting for Merck, unrelated to SGLT2 inhibitors.