Impact of diabetes on the effects of sodium glucose co-transporter-2 inhibitors on kidney outcomes: collaborative meta-analysis of large placebo-controlled trials

Abstract

Background: Large trials have shown that sodium glucose co-transporter-2 (SGLT2) inhibitors reduce the risk of adverse kidney and cardiovascular outcomes in patients with heart failure or chronic kidney disease, or with type 2 diabetes and high risk of atherosclerotic cardiovascular disease. None of the trials recruiting patients with and without diabetes were designed to assess outcomes separately in patients without diabetes.

Methods: We did a systematic review and meta-analysis of SGLT2 inhibitor trials. We searched the MEDLINE and Embase databases for trials published from database inception to Sept 5, 2022. SGLT2 inhibitor trials that were double-blind, placebo-controlled, performed in adults (age ≥18 years), large (≥500 participants per group), and at least 6 months in duration were included. Summary-level data used for analysis were extracted from published reports or provided by trial investigators, and inverse-variance-weighted meta-analyses were conducted to estimate treatment effects. The main efficacy outcomes were kidney disease progression (standardised to a definition of a sustained ≥50% decrease in estimated glomerular filtration rate [eGFR] from randomisation, a sustained low eGFR, end-stage kidney disease, or death from kidney failure), acute kidney injury, and a composite of cardiovascular death or hospitalisation for heart failure. Other outcomes were death from cardiovascular and non-cardiovascular disease considered separately, and the main safety outcomes were ketoacidosis and lower limb amputation. This study is registered with PROSPERO, CRD42022351618.

Findings: We identified 13 trials involving 90 413 participants. After exclusion of four participants with uncertain diabetes status, we analysed 90 409 participants (74 804 [82·7%] participants with diabetes [>99% with type 2 diabetes] and 15 605 [17·3%] without diabetes; trial-level mean baseline eGFR range 37-85 mL/min per 1·73 m²). Compared with placebo, allocation to an SGLT2 inhibitor reduced the risk of kidney disease progression by 37% (relative risk [RR] 0·63, 95% CI 0·58-0·69) with similar RRs in patients with and without diabetes. In the four chronic kidney disease trials, RRs were similar irrespective of primary kidney diagnosis. SGLT2 inhibitors reduced the risk of acute kidney injury by 23% (0·77, 0·70-0·84) and the risk of cardiovascular death or hospitalisation for heart failure by 23% (0·77, 0·74-0·81), again with similar effects in those with and without diabetes. SGLT2 inhibitors also reduced the risk of cardiovascular death (0·86, 0·81-0·92) but did not significantly reduce the risk of non-cardiovascular death (0·94, 0·88-1·02). For these mortality outcomes, RRs were similar in patients with and without diabetes. For all outcomes, results were broadly similar irrespective of trial mean baseline eGFR. Based on estimates of absolute effects, the absolute benefits of SGLT2 inhibition outweighed any serious hazards of ketoacidosis or amputation.

Interpretation: In addition to the established cardiovascular benefits of SGLT2 inhibitors, the randomised data support their use for modifying risk of kidney disease progression and acute kidney injury, not only in patients with type 2 diabetes at high cardiovascular risk, but also in patients with chronic kidney disease or heart failure irrespective of diabetes status, primary kidney disease, or kidney function.

Funding: UK Medical Research Council and Kidney Research UK.

Figure 1

Effect of sodium glucose co-transporter-2 inhibition on kidney disease outcomes by diabetes status Kidney disease progression was defined as a sustained decrease in eGFR (≥50%) from randomisation, a sustained low eGFR, end-stage kidney disease, or death from kidney failure in all presented trials. Outcome definition details for each trial are provided in the appendix (pp 9–11). Rate values are not presented for the combined subtotal and total populations due to the heterogeneity in rates across the individual trials. eGFR=estimated glomerular filtration rate. RR=relative risk. SGLT2i=sodium glucose co-transporter-2 inhibitor. NA=not available. *One participant without diabetes in DELIVER was missing a baseline creatinine measurement and was excluded.

Figure 2

Effect of sodium glucose co-transporter-2 inhibition on kidney disease progression by presumed primary kidney disease (chronic kidney disease trials only) Effects in IgA nephropathy, focal segmental glomerulosclerosis, and other glomerular diseases considered separately are provided in the appendix (p 18). Rate values are not presented for the combined subtotal and total populations due to the heterogeneity in rates across the individual trials. eGFR=estimated glomerular filtration rate. RR=relative risk. SGLT2i=sodium glucose co-transporter-2 inhibitor. *RR in the diabetic kidney disease or nephropathy subgroup excluding SCORED (which did not formally assess primary kidney disease) is 0·59 (95% CI 0·52–0·68).

Figure 3

Effect of sodium glucose co-transporter-2 inhibition on heart failure and mortality outcomes by diabetes status Outcome data sources by trial are provided in the appendix (pp 10–11). Effects on heart failure and mortality were also analysed by trial with event rate per 1000 patient-years presented for each trial (appendix pp 19–20). eGFR=estimated glomerular filtration rate. RR=relative risk. SGLT2i=sodium glucose co-transporter-2 inhibitor. *Cardiovascular death or hospitalisation for heart failure outcomes exclude urgent heart failure visits. †Data from SOLOIST-WHF are included in totals but excluded from the stable heart failure trials group as the trial included patients with recent acute decompensated heart failure.

Figure 4

Effect of sodium glucose co-transporter-2 inhibition on ketoacidosis and lower limb amputation by diabetes status Effects on ketoacidosis and lower limb amputation were also analysed by trial with event rate per 1000 patient-years presented for each trial (appendix pp 21–22). Total values and forest plots are not presented for ketoacidosis due to the small number of events in patients without diabetes. eGFR=estimated glomerular filtration rate. RR=relative risk. SGLT2i=sodium glucose co-transporter-2 inhibitor. *Data from SOLOIST-WHF are included in totals but excluded from the stable heart failure trials group as the trial included patients with acute decompensated heart failure. †The hypothesis that SGLT2i might increase the risk of lower limb amputation was first raised by results from the CANVAS trial; the subtotal excluding CANVAS therefore reflects the combined results from the independent set of hypothesis-testing trials.

Figure 5

Absolute benefits and harms of SGLT2 inhibition per 1000 patient-years by diabetes status and patient group Patient group specific absolute effects estimated by applying the diabetes subgroup specific RR to the average event rate in the placebo arms (first event only). Negative numbers indicate events avoided by SGLT2 inhibition per 1000 patient-years. Error bars represent SE in the numbers of events avoided or caused, estimated from uncertainty in the RRs. Mean eGFR values are given for combined trial populations by patient group and diabetes status. Placebo population mean event rates are the absolute numbers of events per 1000 patient-years in the placebo groups of all trials in the relevant subpopulation. eGFR=estimated glomerular filtration rate. RR=relative risk. SGLT2i=sodium glucose co-transporter-2 inhibitor. *Additionally, two (SE 0·5) fewer myocardial infarctions per 1000 patient-years of SGLT2i treatment were observed in the diabetes and high atherosclerotic cardiovascular risk group. †RRs to determine absolute effects for lower limb amputation included CANVAS. ‡Too few ketoacidosis events to estimate absolute effects.