Over the past several years, SGLT2 inhibitors — drugs originally developed to lower blood sugar in diabetes — have quietly become one of the foundational therapies for heart failure. Trials of empagliflozin and dapagliflozin showed they reduce hospitalizations and death across a broad swath of heart-failure patients, and guidelines now treat them as standard care. But those landmark trials drew a boundary, and on the far side of it sit some of the sickest patients of all: those whose hearts have failed so completely that they now run on a left ventricular assist device (LVAD), an implanted mechanical pump. The big trials largely excluded them, leaving a genuine evidence gap. A University of Chicago Phase 4 study, NCT05278962, set out to start filling it — and its newly posted results are a useful lesson in how to read a small trial.
An LVAD is exactly what it sounds like: a surgically implanted pump that takes over the work of the heart's main pumping chamber, used for patients with severe heart failure with reduced ejection fraction either as a bridge to transplant or as long-term support. These patients are on intensive medication regimens already, and whether adding an SGLT2 inhibitor on top is safe and helpful in the presence of a mechanical pump is a question with essentially no large-trial answer. That is the gap the Chicago team aimed at.
Why would the answer be uncertain in the first place? SGLT2 inhibitors are thought to help heart failure through several mechanisms at once — gentle diuresis that offloads fluid, improved cardiac energy metabolism, and effects on the kidney that ease the heart's workload. In a patient whose failing ventricle has been mechanically bypassed by a pump, it is genuinely unclear how many of those benefits still apply. The pump already handles the circulation the drug would otherwise be trying to protect, so the physiology that made SGLT2 inhibitors so effective in the landmark trials may operate differently here. That uncertainty is precisely why an empirical test, rather than an assumption, is warranted.
The design — and its honest scale
This was a randomized Phase 4 trial, open-label, enrolling 32 adult patients with heart failure with reduced ejection fraction who had LVADs. Participants were randomized to either take an SGLT2 inhibitor — empagliflozin 10 mg daily or dapagliflozin 10 mg daily, depending on which their insurance formulary covered — for six months, or to continue routine heart-failure care without one. Data were drawn from the participants' medical records. The honest framing, which the trial's own structure makes clear, is that this is a small pilot: 32 patients split into two arms means roughly 18 on the drug and 13 without it. A study this size can establish feasibility and safety and hint at a direction; it cannot prove that a drug changes hard outcomes. Reading it as anything more would be a mistake, and the most valuable thing an article can do is say so plainly.
What the evidence shows
The primary endpoint was the change in left ventricular end-diastolic dimension (LVEDD) — a measure of how dilated the heart's main chamber is — at six months. A smaller chamber suggests favorable remodeling, the heart structurally recovering rather than stretching further. The SGLT2 inhibitor arm showed a change of −1.1 and the no-SGLT2i arm −0.9 on the registry's reported measure; the two are close, and with these sample sizes the difference does not establish a clear structural benefit one way or the other.
The secondary endpoints are where the more suggestive signals appear, and they line up with what is known about how these drugs work. SGLT2 inhibitors act partly as mild diuretics, promoting fluid and glucose excretion through the urine, so weight and fluid status are natural places to look. The SGLT2i arm showed a mean weight change of −1.2 kg while the control arm gained +0.7 kg — a modest divergence consistent with the drugs' known fluid-offloading effect. The registry also tracked diuretic dose and ejection fraction as secondary measures, the kinds of physiologic markers a pilot study collects to decide whether a larger trial is worth running.
The safety column is the real takeaway
For a pilot study in a fragile, device-dependent population, the safety data may be the single most important result, and here it is clean: the posted adverse-event tables record zero serious adverse events in both the SGLT2i arm (18 patients) and the no-SGLT2i arm (13 patients) over the six-month window. In a population on long-term mechanical circulatory support, demonstrating that adding an SGLT2 inhibitor did not produce serious harm is itself a meaningful, decision-relevant finding — it is the prerequisite for ever running the larger outcomes trial that could change practice.
Why this trial matters
NCT05278962 is a model of what a good small trial is for: it takes a population the major studies left out, asks whether a now-standard therapy is safe and plausibly beneficial in that group, and generates the safety data and effect-size hints needed to justify — or not — a definitive trial. The results suggest SGLT2 inhibitors are tolerated in LVAD patients and may offer modest weight and fluid benefits, while honestly stopping short of any claim about hard outcomes that 32 patients could never support. For clinicians caring for this small but growing population, it is a first data point where there had been almost none. The full LVEDD, weight, diuretic, and adverse-event tables are posted on the ClinicalTrials.gov study page, which is the authoritative source for the figures here.