GLP-1 drugs linked to fewer nerve complications than DPP-4 inhibitors in diabetes

GLP-1 receptor agonists, the blockbuster class of drugs behind medications such as semaglutide and liraglutide, have become famous for reshaping blood sugar control and body weight in type 2 diabetes. Now a new study suggests they may also change the fate of one of diabetes’ most feared complications: the diabetic foot. Researchers analyzing the medical records of nearly 40,000 U.S. adults with type 2 diabetes and nerve damage report that patients starting a GLP-1 receptor agonist developed diabetic foot ulcers at a lower rate than similar patients starting a DPP-4 inhibitor, another incretin-based diabetes drug. But the analysis, published in the Journal of Neurology, also carries a cautionary note: the GLP-1 group experienced nearly twice the rate of Charcot neuroarthropathy, a rare and destructive collapse of the bones and joints of the foot.

Diabetic peripheral neuropathy, the progressive damage to peripheral nerves caused by chronic high blood sugar, affects roughly half of people with diabetes over their lifetime and sets the stage for a cascade of devastating outcomes. When sensation in the feet is lost, minor injuries go unnoticed, wounds fail to heal, ulcers develop, and infections can burrow into bone. In the worst cases, the resulting osteomyelitis or the fragmented, inflamed joints of Charcot neuroarthropathy end in amputation. The condition is also a marker of systemic vulnerability; patients with neuropathic foot complications face substantially elevated mortality. With GLP-1 receptor agonists now prescribed to millions of people worldwide, understanding how these drugs influence the neuropathic foot has become a pressing clinical question.

Led by Fady Tawfik of Howard University College of Medicine, with colleagues at Texas A&M University College of Medicine and the University of Maryland School of Medicine, the research team turned to the TriNetX US Collaborative Network, a federated database aggregating de-identified electronic health records from dozens of American healthcare organizations. They identified adults with a diagnosis of type 2 diabetes mellitus and diabetic neuropathy, unspecified, coded under ICD-10-CM as E11.40, who had newly initiated either a GLP-1 receptor agonist or a DPP-4 inhibitor. Both drug classes act on the incretin system—the hormonal axis that amplifies insulin secretion after meals—but through different mechanisms. GLP-1 receptor agonists mimic the glucagon-like peptide-1 hormone directly and resist enzymatic breakdown, while DPP-4 inhibitors block the enzyme dipeptidyl peptidase-4 that normally degrades endogenous GLP-1, modestly prolonging its action.

Because patients prescribed these drugs often differ systematically in age, weight, kidney function, and overall disease burden, naive comparisons would be misleading. The investigators therefore applied 1:1 propensity score matching, a statistical technique that pairs each GLP-1 receptor agonist user with a DPP-4 inhibitor user who shares a similar demographic and clinical profile, including comorbidities and concomitant medications. After exclusions and matching, the final cohorts contained 19,770 patients per group, well balanced on the measured baseline covariates. The researchers then tracked five outcomes over one and two years: diabetic foot ulcer, lower-extremity amputation, osteomyelitis of the foot or ankle, Charcot neuroarthropathy, and all-cause mortality. Time-to-event analyses employed Kaplan–Meier curves and Cox proportional hazards models, with a Bonferroni correction setting the threshold for statistical significance at p less than 0.01 to guard against false positives across multiple comparisons.

The headline result concerned foot ulcers, the most common entry point into the cycle of neuropathic foot disease. At one year, 2.2 percent of GLP-1 receptor agonist users had been diagnosed with a diabetic foot ulcer compared with 2.7 percent of DPP-4 inhibitor users, corresponding to a hazard ratio of 0.813 with a 95 percent confidence interval of 0.716 to 0.922. In practical terms, the GLP-1 group experienced roughly a 19 percent relative reduction in ulcer risk over the first year. All-cause mortality was also lower in the GLP-1 group, a finding the authors interpreted as hypothesis-generating rather than definitive, given the observational design and the possibility of residual confounding.

Not every result favored the newer drugs. The apparent survival advantage and ulcer reduction came alongside an unexpected signal: Charcot neuroarthropathy occurred in 0.3 percent of GLP-1 users versus 0.2 percent of DPP-4 inhibitor users, a hazard ratio of 1.993 with a confidence interval of 1.297 to 3.064, meaning nearly a doubling of risk that reached statistical significance. By contrast, rates of lower-extremity amputation, at 0.5 percent in both groups, and foot or ankle osteomyelitis, at 0.4 percent in both groups, were indistinguishable, with hazard ratios crossing unity comfortably.

Why might a drug that reduces foot ulcers simultaneously raise the risk of Charcot neuroarthropathy? The authors and prior literature point to several plausible mechanisms. Charcot neuroarthropathy is widely understood as an “imperfect storm”: an injury to an insensate foot triggers an exaggerated inflammatory response in which osteoclast-driven bone resorption outpaces repair, producing fractures, joint disorganization, and the classic rocker-bottom deformity. Rapid improvements in glycemic control are a recognized trigger for acute neuropathic and Charcot events, a phenomenon related to treatment-induced neuropathy of diabetes, in which abruptly normalized blood sugar provokes acute painful neuropathy and arterio-venous shunting in nerve and bone microvasculature. GLP-1 receptor agonists are potent glucose-lowering agents, and their initiation in patients with poorly controlled diabetes could reproduce this dynamic, destabilizing bone in a foot already numbed by neuropathy.

Counteracting mechanisms may explain the ulcer benefit. Experimental work has shown that GLP-1 receptor signaling exerts neuroprotective and anti-inflammatory effects on peripheral nerves. Studies in diabetic rodent models demonstrated that the GLP-1 receptor agonist exendin-4 improved experimental polyneuropathy, and research in streptozotocin-induced diabetic rats found that GLP-1 receptor activation ameliorated nerve dysfunction by damping the p38 MAPK and nuclear factor kappa-B inflammatory pathways. Human imaging studies have reported that GLP-1 receptor agonists reverse nerve morphological abnormalities in diabetic peripheral neuropathy, and a recent meta-analysis in the Journal of Neurochemistry concluded that the class shows favorable effects on diabetic peripheral neuropathy overall. Independent of nerves, GLP-1 signaling appears to accelerate wound repair: liraglutide facilitated keratinocyte migration and healing through the PI3K/Akt pathway, exendin-4 accelerated diabetic wound closure in surgical models, and recent reviews describe GLP-1 receptor agonists as emerging modulators of inflammation and angiogenesis in chronic wounds, partly through vascular endothelial growth factor signaling. Fewer ulcers and better healing would naturally translate into fewer deep infections, which is consistent with the similar osteomyelitis rates despite different ulcer trajectories.

The findings land in a contentious literature. Previous emulated target trials have produced divergent results: one published in Diabetes Care comparing GLP-1 receptor agonists with SGLT2 inhibitors found differential amputation outcomes between classes, while another in Annals of Internal Medicine examined sodium-glucose cotransporter-2 inhibitors against GLP-1 receptor agonists for diabetic foot disease with mixed conclusions. A pharmacological database study in Diabetes, Obesity and Metabolism examined incretin-based therapy and foot ulcer risk broadly, and a nationwide observational study in Diabetes Care reported that GLP-1 receptor agonists were associated with reduced mortality after diabetic foot ulcers. The new Journal of Neurology analysis is among the first to focus specifically on patients who already have neuropathy—the population at highest risk—and the first in this context to surface Charcot neuroarthropathy as a comparative safety signal between incretin classes.

The authors are careful about causality. As a retrospective cohort study of administrative coding data, the analysis cannot prove that GLP-1 receptor agonists caused fewer ulcers or more Charcot feet. Unmeasured confounding remains possible despite propensity matching; physicians may preferentially prescribe GLP-1 receptor agonists to patients deemed healthier or more adherent, a phenomenon known as channeling bias that has been documented in glucose-lowering drug studies. Coding of Charcot neuroarthropathy is uncommon and inconsistently applied, and although the doubling of risk reached the corrected significance threshold, the absolute numbers were small—roughly three additional cases per thousand patients per year. The modest absolute reduction in ulcers, about five additional ulcer-free patients per thousand at one year, likewise requires context.

Even so, the study carries practical messages. For the growing population of patients with type 2 diabetes and established neuropathy, GLP-1 receptor agonist therapy appears, on balance, favorable for foot health, reinforcing their established cardiovascular and mortality benefits rather than undermining them. But clinicians initiating these potent glucose-lowering agents in patients with preexisting nerve damage should remain alert to the rare possibility of rapid glycemic improvement triggering neuropathic worsening or Charcot joint destruction. The authors emphasize that vigilant foot surveillance during therapy, including prompt evaluation of warmth, swelling, or deformity in an insensate foot, is warranted. Charcot neuroarthropathy caught early can be treated with offloading and immobilization before irreversible deformity sets in; caught late, it is a leading cause of amputation. As GLP-1 receptor agonists continue their meteoric rise from injectable diabetes drugs to near-universal metabolic therapy, this study is a reminder that even celebrated drugs demand close study of their effects on the body’s most vulnerable territories—and that the diabetic foot, more than most, keeps score.