Welcome back. In this issue, three signals from the last few weeks that I think matter — one on the tension between wound healing and kidney protection, one on a therapy finally reaching a population that has waited thirty years, and one from India on what it takes to bring screening to where patients actually are. Plus a voice from the field, from a surgeon who has spent his career watching total contact casting heal ulcers that other approaches could not.
This issue also lands the week after our first virtual Forum on Diabetes Innovation and Limb Preservation. The conversations during the "From Discovery to Patient" panel stayed with me — and, from what several of our young awardees told me afterwards, with them too. I'll come back to it in a future issue.
— Nadège
In this Issue
A prospective propensity score-matched cohort study in *Frontiers in Endocrinology* (June 2026) enrolled 247 adults with type 2 diabetes, diabetic kidney disease (eGFR 30–90), and active wounds requiring medical management at a tertiary wound care center. After matching on 15 baseline covariates, 102 SGLT2 inhibitor users were compared with 102 controls. Median time to complete wound healing was 33 versus 34 days (adjusted HR 1.08, 95% CI 0.82–1.42, p=0.594), and infection rates did not differ. Over the six-month follow-up, no renal progression events occurred in SGLT2 inhibitor users versus 5.9% in controls (p=0.029), and eGFR decline was less than half as steep (−2.3 vs −5.5 mL/min/1.73m², p<0.001).
The reflex to withhold SGLT2 inhibitors when a patient has an open wound dates back to the CANVAS amputation signal of 2017 — a signal that never replicated, and whose regulatory warning was lifted in 2020. Yet the practice persists, at real cost to the patient's kidneys. This is a single-center prospective cohort, not a randomized trial, and the follow-up is short. But it directly addresses a therapeutic hesitation that many of us have inherited without re-examining.
The FINE-ONE trial, published in the *New England Journal of Medicine* in March 2026 (Heerspink et al.), randomized 242 adults with type 1 diabetes and chronic kidney disease across 80 sites in 9 countries to finerenone 10 or 20 mg daily or placebo, on top of standard care. Over six months, urine albumin-to-creatinine ratio dropped by approximately 25% versus placebo (least-squares mean change 0.75, 95% CI 0.65–0.87, p<0.001). On May 21, 2026, the FDA granted Priority Review to Bayer's supplemental NDA. If approved, finerenone would become the first mineralocorticoid receptor antagonist indicated for T1D-associated CKD.
The framing that matters: this is the first investigational therapy in thirty years to show a positive phase 3 result for CKD in type 1 diabetes. The T1D community has watched every major landmark kidney trial exclude them. UACR reduction is a surrogate, not a hard outcome, and hyperkalemia was more common with finerenone (10.1% vs 3.3%) — the usual caveat with this class. But for endocrinologists and nephrologists managing T1D patients whose kidneys are declining despite RAAS blockade, this is the first new tool since the 1990s.
A pragmatic diagnostic accuracy study published in *BMJ Open* (March 2026) evaluated three diabetic retinopathy screening models in rural Punjab, India: non-ophthalmologist-led screening at Health and Wellness Centres, offline AI-assisted smartphone-based screening in community sites, and standard referral-based care. Six hundred people with diabetes aged ≥30 years were enrolled, with retinal images independently graded by masked human graders and adjudicated by a retina specialist. The finding worth reading was not the headline "AI does well" — it was the nuanced one: non-ophthalmologist-led screening at primary care centers meaningfully expanded access, while offline AI showed promise but remained limited by image quality and binary classification outputs. The authors concluded that integrating both approaches, rather than choosing between them, may strengthen coverage in resource-limited settings.
India needs to screen roughly 125 million people with diabetes annually for retinopathy, with one ophthalmologist for every 100,000 patients. That is the arithmetic that reshapes every technical discussion of sensitivity and specificity. The signal here is less about AI performance in isolation than about how task-shifting, technology, and specialist grading fit together when the specialist pipeline cannot expand fast enough.
This month's From the field comes from Frank Ross, general surgeon and director of the Helen L. and Martin S. Kimmel Hyperbaric and Advanced Wound Healing Center at NYU Langone Health. Frank's team just published a paper in Advances in Skin & Wound Care on total contact casting after partial-foot amputations. I asked him to write a few lines on what sits behind the work.
Still amazed: why so few patients get total contact casting
Throughout my career as a general surgeon, I have watched surgery evolve in extraordinary ways. Mesh transformed hernia repair. Laparoscopic surgery changed the way we approached the abdomen. Endovascular techniques revolutionized the treatment of vascular disease. These innovations have fundamentally changed the way we care for patients. Yet when I teach medical students, I often tell them that one of the most remarkable treatments I have ever witnessed is not a new operation or an expensive technology. It is total contact casting.
Even after years of caring for patients with diabetic foot ulcers, it never ceases to amaze me. I have seen large diabetic foot ulcers that linger for months, sometimes after multiple dressings, advanced wound products, antibiotics, and debridement, begin to close within weeks once pressure is properly offloaded. Watching the body heal itself after simply eliminating the mechanical forces preventing healing is one of the most satisfying experiences in surgery.
Which makes me continually ask a simple question: if total contact casting works this well, why do so few patients ever receive it?
The answer is not a lack of science. The problem is implementation. Total contact casting is a specialized skill. It requires training, experience, time, and a dedicated team. Few clinicians receive meaningful hands-on instruction, and many wound centers simply do not have the personnel or expertise to offer it consistently. As a result, patients often receive treatments that are easier to deliver rather than those that are most effective.
As physicians and scientists, we should absolutely continue searching for a cure for diabetes. We should pursue better medications, regenerative therapies, artificial intelligence, and every innovation that holds promise for the future. But while we search for tomorrow's breakthroughs, we must not overlook what we can do today. We already possess the knowledge and tools to heal many diabetic foot ulcers and prevent countless amputations. Our greatest challenge is no longer discovering what works: it is ensuring that every patient has access to it. I remain optimistic.
— Frank Ross, MD
Director, Helen L. and Martin S. Kimmel Hyperbaric and Advanced Wound Healing Center
NYU Langone Health
OneGuild Pulse is a free monthly publication of OneGuild Institute. Forward freely.
Subscribe at oneguild.org/pulse. Find us on Instagram at @diabetespulse.
— Nadège Sheehan, PhD
President, OneGuild Institute
New York, July 2026
Read the June letter · All letters
