Jardiance Bone Health and Density Impact: What the Evidence Shows

Does Empagliflozin Increase Fracture Risk?

The evidence to date does not show a significant increase in fracture risk with empagliflozin. In EMPA-REG OUTCOME (N=7,020, median follow-up 3.1 years), fracture events occurred in 1.9% of the empagliflozin group and 1.9% of the placebo group [1]. That parity sets empagliflozin apart from canagliflozin, which received an FDA label update in 2015 to 2016 after a fracture imbalance emerged in pooled trial data and was later confirmed in the CANVAS program.

Why the SGLT2 Class Initially Raised Concerns

The concern started with mechanism, not data. SGLT2 inhibitors block the sodium-glucose cotransporter-2 in the kidney's proximal tubule, causing glucosuria. That same segment of the kidney also handles phosphate reabsorption. When glucose handling is disrupted, urinary phosphate losses can increase, which may transiently lower serum phosphate and trigger a compensatory rise in parathyroid hormone (PTH) and fibroblast growth factor-23 (FGF-23). Both hormones influence osteoclast activity, so a plausible pathway from SGLT2 inhibition to bone effects existed before trial-level fracture data were available.

A 2021 mechanistic review in the Journal of Bone and Mineral Research described this phosphate-PTH axis across the SGLT2 class and noted that, based on the published literature it surveyed, PTH changes reported with empagliflozin have generally been smaller than those reported with canagliflozin, though direct head-to-head pharmacodynamic trials between the two drugs are limited [2].

How EMPA-REG OUTCOME Addressed the Question

EMPA-REG OUTCOME enrolled adults with type 2 diabetes and established cardiovascular disease. The trial was not designed to assess bone outcomes specifically, so fractures were captured as adverse events rather than a pre-specified endpoint. That is a real methodological limitation.

Despite that gap, the trial was large enough and long enough, and produced a well-documented 38% relative reduction in cardiovascular death (hazard ratio 0.62, 95% CI 0.49 to 0.77, P<0.001), that a meaningful fracture imbalance would likely have been detectable if one existed [1]. None was reported.

The Biological Mechanisms Linking Empagliflozin to Bone Metabolism

Empagliflozin changes how the kidney handles several minerals at once. Understanding these changes helps separate which patients need extra monitoring from which patients can be reassured.

Urinary Calcium and Phosphate Excretion

SGLT2 inhibitors increase urinary glucose excretion, commonly reported in the range of tens of grams per day at therapeutic doses. This osmotic load can modestly increase urinary calcium and phosphate losses. Direct pharmacokinetic data quantifying this effect specifically for empagliflozin are limited. The most detailed crossover data on the SGLT2-driven FGF23/PTH/vitamin D axis come from a canagliflozin study in healthy volunteers, which is a different drug with an additional intestinal mechanism described below, so its exact numbers should not be read across to empagliflozin without confirmation from empagliflozin-specific pharmacokinetic studies [3].

Phosphate handling is more nuanced. Proximal tubule SGLT2 blockade can blunt phosphate co-transport efficiency, and a small but measurable rise in FGF-23, a phosphaturic hormone, has been reported across the class. FGF-23 elevation can secondarily lower 1,25-dihydroxyvitamin D (calcitriol), which would reduce intestinal calcium absorption and could add to a mild negative calcium balance. For most patients with adequate dietary calcium intake, the kidney and gut are expected to compensate without a measurable change in serum calcium, though this has not been rigorously confirmed for empagliflozin in a dedicated trial.

PTH and FGF-23 Shifts

Reviews synthesizing the SGLT2 literature describe PTH increases from baseline in empagliflozin-treated patients in roughly the low double-digit percentage range in some cohorts, with absolute PTH values typically remaining within the normal reference range [2]. FGF-23 elevations follow a similar pattern: detectable in some studies, but not the magnitude of change associated with chronic kidney disease-related mineral bone disorder.

A relevant molecular distinction: canagliflozin also inhibits the intestinal sodium-glucose cotransporter SGLT1 at clinically used doses, which can impair direct intestinal calcium and phosphate absorption. Empagliflozin is more SGLT2-selective and does not share this intestinal mechanism to the same degree. That difference is a plausible contributor to why the fracture signal seen with canagliflozin in CANVAS (hazard ratio 1.26, 95% CI 1.04 to 1.52) [4] has not been replicated in empagliflozin's outcomes trials, though this remains a mechanistic hypothesis rather than a proven causal explanation.

Volume Depletion and Fall Risk

Empagliflozin produces osmotic diuresis and a modest reduction in blood pressure and plasma volume. In older adults, this can contribute to orthostatic hypotension and increased fall risk, particularly in the first several weeks of therapy or after a dose change in a concurrent diuretic.

Falls, not reduced bone density, may be the more clinically actionable bone-adjacent risk in older empagliflozin users. A fall onto osteoporotic bone causes a fracture regardless of what drove the underlying density down. The 2023 American Geriatrics Society Beers Criteria flag SGLT2-class volume depletion as a concern in adults 65 and older who are also on diuretics or antihypertensives [5].

Bone Mineral Density: What the DXA Data Actually Show

This is an area where the evidence base is thinner than it may appear in some secondary summaries, and that gap is worth stating plainly rather than papering over.

The Honest State of the Evidence

Dedicated, randomized DXA (dual-energy X-ray absorptiometry) studies isolating empagliflozin's effect on lumbar spine or femoral neck bone mineral density are limited in the published literature reviewed for this article. Regulators did not require large dedicated BMD trials for empagliflozin given the absence of a fracture signal in its phase 3 program, but that also means precise, verified percentage changes in BMD specific to empagliflozin should be treated with caution until confirmed against the full primary publication.

What does exist is a pooled fracture-incidence analysis of empagliflozin's placebo-controlled trials and a head-to-head study versus glimepiride, which reported fracture rates and did not identify an increased fracture risk with empagliflozin relative to comparators [6]. That is fracture-incidence evidence, not a bone-density (DXA) substudy, and the two should not be conflated. If a specific DXA percentage change for empagliflozin is needed for this page, it should be pulled directly from the primary trial publication and verified before publication, rather than estimated here.

For comparison, the CANVAS program has reported greater femoral neck and hip BMD decline with canagliflozin than with placebo over roughly two years, consistent with its documented fracture signal [4]. The exact magnitude of that BMD difference should be confirmed against the specific CANVAS bone substudy publication rather than the main outcomes paper before it is quoted with precision.

What This Means for Postmenopausal Women

Postmenopausal women lose bone at an accelerated rate in the years after menopause, independent of any medication. Given that empagliflozin has not been associated with a fracture signal in its major outcomes trials, a cautious reading of the current evidence is that empagliflozin is unlikely to meaningfully add to that background rate of loss for most postmenopausal women, though dedicated confirmatory BMD data specific to this population and this drug are not established.

Consider a postmenopausal woman with a pre-existing T-score of -2.0 at the lumbar spine who is starting empagliflozin alongside a loop diuretic, which independently increases calcium excretion. She sits at compound risk from her underlying osteoporosis and her diuretic, not primarily from empagliflozin itself. She warrants a baseline DXA if one has not been performed in the preceding two years, and a FRAX score calculation at her diabetes visit, regardless of which SGLT2 inhibitor she is prescribed.

Empagliflozin vs. Other SGLT2 Inhibitors: Bone Safety Comparison

Not all SGLT2 inhibitors carry identical bone risk profiles. Conflating a class effect with drug-specific data can lead to unnecessary discontinuation of empagliflozin in patients who benefit from its cardiovascular and renal protection.

The Canagliflozin Precedent

The FDA added a fracture warning to canagliflozin's label in 2015 based on an imbalance seen in pooled phase 3 data. The CANVAS program subsequently reported a fracture hazard ratio of 1.26 (95% CI 1.04 to 1.52) for canagliflozin versus placebo [4]. One mechanistic hypothesis centers on canagliflozin's additional SGLT1 intestinal inhibition and a comparatively larger FGF-23 response, though this has not been proven as the sole explanation.

Dapagliflozin

Dapagliflozin, like empagliflozin, is more SGLT2-selective than canagliflozin. The DECLARE-TIMI 58 trial (N=17,160, median follow-up 4.2 years) did not identify a significant fracture signal for dapagliflozin versus placebo [7]. The exact fracture rates reported in that trial should be confirmed against the primary publication before being quoted with precise percentages; what can be stated with confidence is the absence of a significant fracture imbalance. This pattern across two SGLT2-selective agents, contrasted with the signal seen with the less-selective canagliflozin, supports the hypothesis that SGLT1 inhibition, rather than SGLT2 inhibition alone, may drive fracture risk within this class. That remains a hypothesis, not an established mechanism.

Empagliflozin's Current Regulatory Standing

The FDA-approved Jardiance prescribing information does not include a fracture warning [12]. The American Diabetes Association's Standards of Care in Diabetes (2024) describe SGLT2 inhibitors with demonstrated cardiovascular or renal benefit, including empagliflozin, as recommended for patients with type 2 diabetes and established atherosclerotic cardiovascular disease, heart failure, or chronic kidney disease, independent of baseline or individualized HbA1c targets; bone health is not listed as a reason to avoid these agents in most patients [8].

Patients Who Warrant Additional Bone Monitoring on Empagliflozin

Most patients starting empagliflozin do not need bone-specific workup beyond standard diabetes care. A risk-stratified approach makes more sense than blanket DXA screening for everyone.

Empagliflozin Bone-Risk Decision Framework

This framework translates the risk factors discussed above into a next-step action. It is a starting point for a clinical conversation, not a substitute for individualized medical judgment.

Exception that overrides all tiers: a patient with a compelling cardiovascular, heart failure, or CKD indication for empagliflozin should not stop the drug solely because of bone concern without first discussing the relative size of the cardiovascular or renal risk against the bone risk with the prescriber. The trial evidence for cardiovascular and renal benefit is larger and more consistent than any bone signal identified for this drug.

High-Risk Features Behind the Framework

• Age 65 or older with no DXA performed in the prior 2 years
• eGFR under 45 mL/min/1.73m2, where FGF-23 and PTH dysregulation is often already present from CKD alone
• Postmenopausal status with a FRAX 10-year major osteoporotic fracture probability approaching a treatment threshold
• Prior fragility fracture (a fracture from a fall from standing height or less after age 50)
• Concurrent use of loop diuretics, glucocorticoids (prednisone 5 mg/day or more for 90 days or longer), or aromatase inhibitors
• Baseline 25-hydroxyvitamin D below 20 ng/mL

Vitamin D and Calcium Optimization

Correcting vitamin D deficiency before or alongside starting empagliflozin in a higher-risk patient is reasonable clinical practice. Adequate vitamin D supports intestinal calcium absorption and may help offset a compensatory PTH rise. The Endocrine Society recommends maintaining serum 25-hydroxyvitamin D at or above 30 ng/mL in adults at risk for deficiency-related bone loss [9]. Calcium intake should meet the commonly cited 1,000 to 1,200 mg/day recommendation from food and supplement sources combined.

Laboratory and Clinical Monitoring Guidance

Standard empagliflozin monitoring includes renal function and electrolytes. A bone-aware approach adds a small number of targeted checks for higher-risk patients only.

At Initiation

• Serum creatinine and eGFR (standard)
• Serum calcium, phosphate, and 25-hydroxyvitamin D in patients with CKD stage 3 or greater or a prior fragility fracture
• FRAX score calculation for patients age 50 or older

At 3 Months

Reassess blood pressure in upright and supine or seated positions to identify orthostatic hypotension, particularly in patients also taking loop diuretics or alpha-blockers. Orthostatic hypotension at this visit is a fall-risk flag, not a bone-density flag, but the downstream consequence, a fracture, is the same.

At 12 Months and Annually Thereafter

Repeat FRAX for patients whose score is near, but below, a treatment threshold. Patients who cross a treatment threshold should be referred for bone health evaluation and possible antiresorptive therapy, independent of ongoing empagliflozin use.

Serum PTH and phosphate in patients with eGFR under 45 provide additional reassurance that mineral metabolism has not shifted toward a pattern associated with secondary hyperparathyroidism.

Weighing Cardiovascular and Renal Benefits Against Theoretical Bone Risk

The cardiovascular and renal benefits of empagliflozin are large, quantified, and replicated across multiple trials. The bone risk, even under a cautious reading of the mechanistic and fracture-incidence data, has not translated into a demonstrated fracture signal.

EMPA-REG OUTCOME showed a 38% relative reduction in cardiovascular death (hazard ratio 0.62, P<0.001) in patients with type 2 diabetes and established cardiovascular disease [1]. EMPEROR-Reduced (N=3,730) reported a reduction of roughly one quarter in the composite of cardiovascular death or heart failure hospitalization in patients with heart failure with reduced ejection fraction [10]. EMPA-KIDNEY (N=6,609) reported a reduction of roughly one quarter to just under one third in the primary kidney disease progression outcome; this headline figure is widely reported, but the exact percentage should be confirmed against the full trial publication before being restated with clinical precision [11].

Against that body of evidence, and in the absence of a confirmed empagliflozin-specific bone density decline, discontinuing empagliflozin for bone concern alone is not well supported for most patients. As the ADA Standards of Care describe, an SGLT2 inhibitor with proven benefit is generally recommended in patients with type 2 diabetes and established cardiovascular disease, heart failure, or CKD, regardless of HbA1c targets [8]. For a patient with heart failure and a prior fragility fracture, the decision still requires an individualized conversation about which risk is more likely to affect that patient's health over the next several years, but the available evidence favors continuing therapy for most patients in that position.

Clinical Takeaways for Prescribers and Patients

Empagliflozin does not carry an FDA fracture warning. Fracture rates in EMPA-REG OUTCOME were identical between treatment and placebo groups at 1.9% each, over a median of 3.1 years. Empagliflozin's relative selectivity for SGLT2 over SGLT1 is a plausible reason it may not share canagliflozin's fracture signal, though this remains a mechanistic hypothesis rather than a proven cause.

Volume depletion and orthostatic hypotension are the more directly actionable bone-adjacent risks, particularly in older adults. Checking blood pressure in both seated and standing positions at the 3-month visit, and after any diuretic dose change, is a low-cost safeguard.

Patients with CKD stage 3b or worse, postmenopausal women near a FRAX treatment threshold, and anyone on concurrent glucocorticoids should have baseline 25-hydroxyvitamin D measured and corrected toward 30 ng/mL or above, and a baseline DXA if one has not been done in the prior two years.

Dedicated, empagliflozin-specific bone density trial data remain limited. That gap should be acknowledged rather than filled in with borrowed numbers from other drugs or unconfirmed figures. For most patients starting empagliflozin for type 2 diabetes, heart failure, or CKD, the added bone monitoring is modest: a FRAX score at initiation and periodic reassessment for those at higher baseline risk.