SGLT-2i as Compared with other Antidiabetic Drugs, does not Increase the Risk of Lower Limb Amputation in T2DM Patients

18 May, 21

Introduction

Treatment with sodium glucose cotransporter-2 inhibitor (SGLT-2i) in type-2 diabetes mellitus (T2DM) patients is associated with cardiorenal benefits. However, there is uncertainty regarding the association between SGLT-2i treatment and risk of lower limb amputation (LLA). Ischemia, especially peripheral artery disease (PAD), is associated with a dramatic increase risk of LLA. Nevertheless, most of the studies with SGLT-2i included T2DM patients with long diabetes duration and high cardiovascular (CV) risk.

Aims

To elucidate the temporal pattern of amputations in patients with T2DM
To determine the risk of amputations as per the use of new and older anti-diabetic drugs (ADDs)
To determine the impact of PAD on the association of therapy and amputation risk

Patient Profile

T2DM patients (age; 18-80 years) were categorized as follows based on the diabetes treatment:

‘SGLT-2i’-exposed to SGLT-2i with or without use of glucagon-like peptide 1 receptor agonist (GLP-1RA) and dipeptidyl peptidase-4 inhibitor (DPP-4i)
‘GLP-1RA’-exposed to GLP-1RA but never exposed to SGLT-2i nor DPP-4i
‘DPP-4i’-exposed to DPP-4i but never exposed to SGLT-2i nor GLP-1RA
‘other ADD’-exposed to an ADD other than SGLT-2i, GLP-1RA, or DPP-4i
‘no ADD’-never initiated ADD

Methods

Study Design

A retrospective analysis

Outcomes

Any LLA or PAD event

Results

A total of 32,93,983 patients with T2DM were identified using the Electronic Medical Records from the USA. Of these, 1,69,739 were SGLT-2i-exposed (SGLT-2i; no exposure to incretins); 1,49,826 were GLP-1RA-exposed (GLP-1RA, no SGLT-2i or DPP-4i exposure); 4,48,225 were DPP-4i-exposed (no exposure to GLP-1RA or SGLT-2i); and 19,54,353 were other ADDs-exposed.
The incidence of amputations per 10 000 adults ranged between 4.7 and 6.8 during 2000-08 and increased significantly to 12.3 in 2017.
During 1,72,11,719 person-years of follow-up post T2DM diagnosis,the rates per 1000 person-years of any amputation and LLAs were similar between SGLT-2i and incretins [95% confidence interval (CI) range: 1.06-1.67], and significantly higher in other groups (95% CI range: 1.96-2.29).
As per propensity score-adjusted pairwise analyses,the risk of LLA was not higher in SGLT-2i vs. GLP1-RA, and was lower in SGLT-2i vs. DPP-4i/other ADD (Table).

Table: The risk of LLA during the study period

Treatment Group	HR (95% CI)
SGLT-2i vs. GLP1-RA	0.88 (0.73, 1.05)
SGLT-2i vs. DPP-4i	0.65 (0.56, 0.75)
SGLT-2i vs. Other ADD	0.43 (0.37, 0.49)

CI: Confidence interval; DPP-4i: Dipeptidyl peptidase-4 inhibitor; HR: Hazard Ratio; SGLT-2i: Sodium glucose cotransporter-2 inhibitor

The rate of LLA did not differ in patients treated with canagliflozin, empagliflozin, or dapagliflozin.
Patients with PAD had more than four-fold higher risk for LLA [range of 95% CI of hazard ratio (HR): 3.6-6.0] or any amputation (range of CI of HR: 3.4-5.7).

Conclusions

The study provides data on more than 3 million T2DM patients from a nationally representative US population.
T2DM patients treated with SGLT-2i did not have increased risk of any amputation or LLA as compared with incretins or other ADDs.
Pre-existing PAD was associated with a greatest increase in the risk of amputation.
The findings are robust enough to put into context any signals seen towards excess amputation in the randomized clinical trials and epidemiological studies of SGLT-2i.

Eur Heart J. 2021 May 7;42(18):1728-1738.