Few drug-safety questions have travelled as directly from a cardiology trial into the podiatry clinic as the relationship between SGLT2 inhibitors and lower-limb amputation. Sodium-glucose co-transporter-2 (SGLT2) inhibitors — canagliflozin, dapagliflozin, empagliflozin and related agents — are now foundational therapy for type 2 diabetes with cardiovascular or kidney disease. Yet a single 2017 trial finding raised the possibility that one of these agents increased the risk of toe and forefoot amputation, and the question has shaped prescribing conversations ever since. Nearly a decade of additional data has clarified the picture considerably, though not entirely.
Where the Concern Originated
The CANVAS Program, reported by Neal and colleagues in the New England Journal of Medicine in 2017, randomised 10,142 participants with type 2 diabetes and elevated cardiovascular risk to canagliflozin or placebo. Canagliflozin reduced major adverse cardiovascular events, but amputations occurred at 6.3 versus 3.4 participants per 1,000 patient-years (hazard ratio 1.97; 95% CI 1.41–2.75). Critically, these were predominantly minor amputations at the level of the toe or metatarsal — precisely the territory of diabetic foot care. The strongest predictor of amputation in the trial was a prior amputation, a finding entirely consistent with what is known about the recurrent nature of diabetic foot disease.
Why the Signal Did Not Replicate
The CREDENCE trial, which studied canagliflozin in 4,401 participants with diabetic nephropathy, found no increase in amputation despite enrolling a population with substantially higher baseline amputation history (5.3% versus 2.3%) and a higher absolute amputation event rate. Arnott and colleagues performed a pooled patient-level analysis of both trials in Diabetes, Obesity and Metabolism (2020) and could not identify any participant characteristic, trial design feature, or foot-care protocol that explained the discrepancy. Their conclusion was measured: given null findings in CREDENCE and in every other large SGLT2 inhibitor trial, the CANVAS result may well have been a chance finding.
Evidence from Meta-Analysis and Real-World Data
Lin and colleagues (Cardiovascular Diabetology, 2021) pooled randomised trial data and found a modestly elevated risk confined to canagliflozin specifically (amputation OR 1.60, 95% CI 1.04–2.46; peripheral arterial disease OR 1.53, 95% CI 1.14–2.05), with no comparable signal across the drug class. Interestingly, their meta-regression found that greater weight reduction and greater blood-pressure reduction on treatment were associated with higher rates of amputation, PAD, and diabetic foot events — raising the possibility that volume contraction and reduced distal perfusion pressure, rather than a direct drug effect, mediate any real risk.
Observational data have been largely reassuring. The OBSERVE-4D study (Ryan et al., 2018) analysed more than 700,000 new users across four US claims databases and found no increased risk of below-knee amputation with canagliflozin compared with non-SGLT2 agents. A meta-analysis of 13 studies involving 2,095,033 patients (Lu and Guo, Therapeutic Advances in Drug Safety, 2023) found no difference in amputation risk between SGLT2 inhibitor and DPP-4 inhibitor users. Comparisons against GLP-1 receptor agonists have been less consistent, and Scheen (Diabetes & Metabolism, 2022) has argued that this apparent difference more likely reflects a protective effect of GLP-1 agonists than harm from SGLT2 inhibitors.
Agent-specific safety analyses reinforce this. An individual participant-level meta-analysis of four large empagliflozin trials encompassing nearly 40,000 patient-years of exposure (Wanner et al., Advances in Therapy, 2024) found no increase in lower-limb amputation. The collaborative meta-analysis of 13 SGLT2 inhibitor trials involving 90,413 participants published in The Lancet in 2022 concluded that the absolute benefits of treatment outweighed any serious hazard of amputation or ketoacidosis.
Regulatory Position and Practical Implications
The US Food and Drug Administration required a boxed warning for canagliflozin in 2017 and removed it on 26 August 2020, citing accumulated trial data and the drug’s demonstrated cardiovascular and renal benefits. The FDA’s communication was explicit that removal of the warning did not eliminate the underlying consideration: clinicians and patients were advised to continue emphasising preventive foot care and to monitor for new pain, tenderness, sores, ulcers, or infection in the legs and feet.
That guidance describes standard diabetic foot care rather than anything specific to this drug class. Patients who warrant SGLT2 inhibition — those with established cardiovascular disease, heart failure, or chronic kidney disease — are by definition patients at elevated foot risk through vascular disease, neuropathy, and renal impairment. Adequate hydration status also deserves attention, given the association between volume and blood-pressure reduction and lower-limb events.
Clinical Summary
The weight of evidence accumulated since 2017 does not support a class-wide increase in amputation risk with SGLT2 inhibitors. The CANVAS signal was not reproduced in CREDENCE, in other large randomised trials, or in large observational cohorts, and no mechanistic explanation for the discrepancy has been established. Any residual risk appears small, possibly confined to canagliflozin, and is outweighed at the population level by substantial cardiovascular and renal benefit. Structured foot surveillance remains appropriate in this population — not because of the medication, but because of the vascular and neuropathic risk profile that led to its prescription.
References
- Neal B, Perkovic V, Mahaffey KW, et al. Canagliflozin and Cardiovascular and Renal Events in Type 2 Diabetes. New England Journal of Medicine. 2017;377(7):644–657. doi:10.1056/NEJMoa1611925
- Arnott C, Huang Y, Neuen BL, et al. The effect of canagliflozin on amputation risk in the CANVAS program and the CREDENCE trial. Diabetes, Obesity and Metabolism. 2020;22(10):1753–1766. doi:10.1111/dom.14091
- Lin C, Zhu X, Cai X, et al. SGLT2 inhibitors and lower limb complications: an updated meta-analysis. Cardiovascular Diabetology. 2021;20(1):91. doi:10.1186/s12933-021-01276-9
- Ryan PB, Buse JB, Schuemie MJ, et al. Comparative effectiveness of canagliflozin, SGLT2 inhibitors and non-SGLT2 inhibitors on the risk of hospitalization for heart failure and amputation in patients with type 2 diabetes mellitus (OBSERVE-4D). Diabetes, Obesity and Metabolism. 2018;20(11):2585–2597. doi:10.1111/dom.13424
- Lu Y, Guo C. Risk of lower limb amputation in diabetic patients using SGLT2 inhibitors, DPP4 inhibitors or GLP-1 agonists: a meta-analysis of 2 million patients. Therapeutic Advances in Drug Safety. 2023;14:20420986231178126. doi:10.1177/20420986231178126
- Scheen AJ. Lower limb amputations: protection with GLP-1 receptor agonists rather than increased risk with SGLT2 inhibitors? Diabetes & Metabolism. 2022;48(2):101325. doi:10.1016/j.diabet.2022.101325
- Nuffield Department of Population Health Renal Studies Group; SGLT2 inhibitor Meta-Analysis Cardio-Renal Trialists’ Consortium. Impact of diabetes on the effects of sodium glucose co-transporter-2 inhibitors on kidney outcomes: collaborative meta-analysis of large placebo-controlled trials. The Lancet. 2022;400(10365):1788–1801. doi:10.1016/S0140-6736(22)02074-8
- Wanner C, Iliev H, Duarte N, et al. Safety of Empagliflozin: An Individual Participant-Level Data Meta-Analysis from Four Large Trials. Advances in Therapy. 2024;41(7):2826–2844. doi:10.1007/s12325-024-02879-w
- US Food and Drug Administration. FDA removes Boxed Warning about risk of leg and foot amputations for the diabetes medicine canagliflozin (Invokana, Invokamet, Invokamet XR). Drug Safety Communication, 26 August 2020.
Comments
