--- canonical_name: Empagliflozin vs. Canagliflozin alternate_names: Jardiance, Invokana, BI 10773, JNJ-28431754, TA-7284 canonical_topic: Empagliflozin vs. Canagliflozin for Health & Longevity short_topic_lc: empagliflozin_vs_canagliflozin creation_date: 2026-0829-1750 creator_ai_fullname: Opus 5 ep_keywords: SGLT2 Inhibitors, Sodium-Glucose Cotransporter 2 Inhibitors, Gliflozins --- # Empagliflozin vs. Canagliflozin for Health & Longevity
Evidence Review created on 08/29/2026 using [AI4L](https://github.com/forever-healthy/AI4L) / Opus 5 **Also known as:** Jardiance, Invokana, BI 10773, JNJ-28431754, TA-7284 ## Motivation Empagliflozin (sold as Jardiance) and canagliflozin (sold as Invokana) are two oral medications that make the kidneys pass sugar out in the urine instead of returning it to the blood. Both were developed for type 2 diabetes, and both were later shown to protect the heart and the kidneys in people who take them over years. Interest in the pair outside diabetes care has two roots. Large outcome studies found that each drug lowered the rate of heart and kidney events, in some cases in people without diabetes at all. Separately, a rigorous animal aging program found that canagliflozin lengthened the lives of male mice — something empagliflozin has never been tested for — which turned the choice between the two into a live question for people taking either drug with long-term health in mind. This review examines what the published evidence shows about how the two differ: how each acts inside the body, which benefits and harms have actually been recorded in people, where their safety signals diverge, and what is known about their practical use. It also marks the places where the evidence is too thin to separate them. **[Benefits](#expected-benefits) - [Risks](#potential-risks--side-effects) - [Protocol](#therapeutic-protocol) - [Conclusion](#conclusion)** ## Recommended Reading This section collects high-level expert commentary and clinical writing that treats these two drugs, or their shared class of SGLT2 inhibitors (sodium-glucose cotransporter 2 inhibitors — medicines that block the kidney protein which reclaims filtered sugar), in substantial depth. - [AMA #53: Metabolic health & pharmacologic interventions: SGLT-2 inhibitors, metformin, GLP-1 agonists, and the impact of statins](https://peterattiamd.com/ama53/) - Peter Attia Attia works through how the SGLT2 inhibitor class that both drugs belong to is used outside diabetes, how it compares with metformin and GLP-1 agonists (injected diabetes drugs), and whether it slows aging. - [Inhibiting a Transporter to Stop Senescence](https://lifespan.io/inhibiting-a-transporter-to-stop-senescence/) - Josh Conway A longevity-focused walkthrough of the experiment showing that canagliflozin clears senescent cells (worn-out cells that linger) through immune surveillance, the leading mechanistic case for a class effect on aging. - [Canagliflozin extends life span in genetically heterogeneous male but not female mice](https://pubmed.ncbi.nlm.nih.gov/32990681/) - Miller et al., 2020 The primary report from the National Institute on Aging's Interventions Testing Program. It is the single piece of evidence that most distinguishes canagliflozin from empagliflozin in longevity terms. - [As We See It: End-Stage Kidney Disease Doubles](https://www.lifeextension.com/magazine/2025/5/end-stage-kidney-disease-doubles) - William Faloon A longevity-audience overview of the SGLT2 inhibitor class in kidney disease, setting the dose ranges of empagliflozin and canagliflozin beside the trial evidence on filtration rate and protein leak. - [Evaluation of the pharmacokinetics, pharmacodynamics and clinical efficacy of empagliflozin for the treatment of type 2 diabetes](https://pubmed.ncbi.nlm.nih.gov/27817207/) - Tomlinson et al., 2017 The fullest account of empagliflozin's handling and dosing, written by independent academics, and useful because it argues explicitly that selectivity differences within the class may not be clinically shared. Note: no directly relevant content was found from Rhonda Patrick, Andrew Huberman or Chris Kresser. Site searches on those platforms returned only passing mentions of the drug class inside articles about diabetes drugs, ketones or kidney disease generally, none of which discuss either drug in depth. Only marginal material was available, so it was not included rather than padding the list. ## Grokipedia [Empagliflozin](https://grokipedia.com/page/Empagliflozin) A dedicated drug page covering empagliflozin's approved uses in diabetes, heart failure and kidney disease, its mechanism, dosing and regulatory history, useful as the reference baseline for the empagliflozin evidence below. [Canagliflozin](https://grokipedia.com/page/Canagliflozin) A dedicated drug page covering canagliflozin's approved uses, dosing, kidney and cardiovascular outcome data, safety signals and regulatory history, useful as the reference baseline against which the comparison below is read. ## Examine No Examine article exists for either empagliflozin or canagliflozin. Both are prescription-only medicines, and Examine.com does not typically cover prescription medications, restricting its coverage to supplements and nutrition topics. ## ConsumerLab No ConsumerLab article exists for either empagliflozin or canagliflozin. Both are prescription-only medicines, and ConsumerLab does not typically cover prescription medications, since its testing program is confined to supplements and food products. ## Systematic Reviews The pooled evidence below covers both what separates the two drugs on effectiveness and what separates them on their principal risks. - [Comparison of Effectiveness Among Different Sodium-Glucose Cotransoporter-2 Inhibitors According to Underlying Conditions: A Network Meta-Analysis of Randomized Controlled Trials](https://pubmed.ncbi.nlm.nih.gov/38293914/) - Kani et al., 2024 Pools 21 placebo-controlled trials and 96,196 participants; finds no consistent efficacy or safety separation between empagliflozin and canagliflozin. - [Comparative Efficacy of Five SGLT2i on Cardiorenal Events: A Network Meta-analysis Based on Ten CVOTs](https://pubmed.ncbi.nlm.nih.gov/34231123/) - Qiu et al., 2022 Ranks both drugs across eight cardiorenal endpoints in ten cardiovascular outcome trials, placing canagliflozin first for major cardiac events and empagliflozin first for death. - [Comparative safety of different sodium-glucose transporter 2 inhibitors in patients with type 2 diabetes: a systematic review and network meta-analysis of randomized controlled trials](https://pubmed.ncbi.nlm.nih.gov/37701900/) - Li et al., 2023 Compares nine agents across 113 trials; reproductive tract infection is the dominant signal, with no separation on amputation, fracture or ketoacidosis (dangerous blood acid build-up). - [Association between sodium-glucose cotransporter 2 (SGLT2) inhibitors and lower extremity amputation: A systematic review and meta-analysis](https://pubmed.ncbi.nlm.nih.gov/32502190/) - Heyward et al., 2020 The central risk comparison: pooled canagliflozin trials show a significant amputation excess, while the empagliflozin trials show none. - [Effects of Anti-Diabetic Drugs on Fracture Risk: A Systematic Review and Network Meta-Analysis](https://pubmed.ncbi.nlm.nih.gov/34721294/) - Zhang et al., 2021 Across 117 trials and 221,364 participants, neither drug significantly altered fracture risk, contradicting the fracture signal seen inside canagliflozin's own program. ## Mechanism of Action Both drugs block SGLT2 in the kidney's proximal tubule, which reclaims most filtered sugar, so glucose and sodium leave in the urine. The extra sodium reaching the macula densa (a salt sensor at the end of that tubule) restores tubuloglomerular feedback (the reflex that narrows the vessel feeding the filter), lowering pressure inside the filtering unit. Osmotic diuresis (extra urine drawn out by sugar), a shift toward fat and ketone fuel, and a rise in red cell mass follow. The two differ in selectivity. Empagliflozin is more than 2,500-fold selective for SGLT2 over SGLT1 (the sugar transporter of the small intestine); canagliflozin is roughly 190-fold, with inhibition constants of 4.0 nM and 770.5 nM, so at 300 mg its gut concentration also slows intestinal sugar uptake ([Ohgaki et al.](https://pubmed.ncbi.nlm.nih.gov/27189972/)). Distribution is narrow: heavy plasma-protein binding keeps both in the circulation, so only the kidney tubule and canagliflozin's gut lumen see meaningful concentrations. Half-life is about 12 hours for empagliflozin ([Tomlinson et al.](https://pubmed.ncbi.nlm.nih.gov/27817207/)) and 10.6 to 13.1 hours for canagliflozin ([Devineni & Polidori](https://pubmed.ncbi.nlm.nih.gov/26041408/)). Neither is a meaningful substrate of the cytochrome P450 enzymes (the liver's main drug-breakdown system); both are cleared by UGT enzymes (which tag a drug with a sugar group for excretion) — empagliflozin via UGT2B7 and UGT1A9, canagliflozin via UGT1A9 and UGT2B4. Why either might touch aging is contested: one account credits blunted post-meal glucose peaks, a second AMPK activation (a cellular fuel gauge) with mTOR suppression (a growth pathway), a third immune clearance of worn-out cells. ## Historical Context & Evolution The lineage begins with phlorizin, isolated from apple tree bark in 1835, which made sugar spill into the urine. For most of the twentieth century that effect was read as a toxicity rather than a therapy, since phlorizin was poorly absorbed and blocked the intestinal transporter too. Canagliflozin was the first of the class approved in the United States, in March 2013; empagliflozin followed in August 2014. Both arrived as glucose-lowering agents with modest effects on weight and blood pressure, and both were required to run cardiovascular safety trials, each paid for by its own maker — Boehringer Ingelheim with Eli Lilly, and Janssen. The [empagliflozin trial](https://pubmed.ncbi.nlm.nih.gov/26378978/) reported in 2015 with lower cardiovascular death and lower death from any cause, unexpected from a glucose-lowering drug. The [canagliflozin program](https://pubmed.ncbi.nlm.nih.gov/28605608/) reported in 2017 with a comparable reduction in cardiovascular events and an unanticipated near-doubling of lower-limb amputation, drawing a boxed warning from the Food and Drug Administration (the United States drug regulator) that year. That warning was withdrawn in August 2020, after [canagliflozin's kidney outcome trial](https://pubmed.ncbi.nlm.nih.gov/30990260/) found no amputation imbalance and later analyses failed to reproduce the signal. The episode is often summarized as the amputation risk having been disproven. More accurately, one large program found it, a second did not, no mechanism was established, and the question closed administratively rather than scientifically. In parallel, a [federally funded mouse aging program](https://pubmed.ncbi.nlm.nih.gov/32990681/) reported in 2020 that canagliflozin extended male lifespan, opening a second line of interest empagliflozin has never been tested against. ## Expected Benefits ### High 🟩 🟩 🟩 #### Reduced Hospitalization for Heart Failure Both drugs cut the rate at which people are admitted for heart failure, and this is the most reliably replicated benefit in the class. The proposed mechanism is lower plasma volume and cardiac preload plus a shift in heart muscle fuel use, not glucose lowering — the effect appears within weeks and holds in people without diabetes. The evidence base is placebo-controlled outcome trials of each agent, all sponsored by their manufacturers, pooled in network meta-analyses. Indirect comparison does not separate the two. **Magnitude:** In a network meta-analysis of ten cardiovascular outcome trials, hazard ratios (HR — how much a treatment changes the rate of an event) for heart failure hospitalization versus placebo were 0.64 (95% confidence interval, or CI — the range within which the true value most likely lies, 0.53–0.77) for canagliflozin and 0.68 (0.59–0.78) for empagliflozin, with fully overlapping intervals ([Qiu et al.](https://pubmed.ncbi.nlm.nih.gov/34231123/)). Trial-level figures: [EMPA-REG OUTCOME](https://pubmed.ncbi.nlm.nih.gov/26378978/) 2.7% versus 4.1%, funded by Boehringer Ingelheim and Eli Lilly; [CANVAS Program](https://pubmed.ncbi.nlm.nih.gov/28605608/), funded by Janssen Research and Development. #### Slowed Progression of Kidney Disease Each drug slows the decline in kidney filtration and reduces the risk of reaching dialysis or transplant. The mechanism is the restored tubuloglomerular feedback described above, which lowers the pressure the filter works against; the benefit is largest where protein leak into the urine is highest. Canagliflozin has the dedicated trial in kidney disease with heavy protein leak; empagliflozin has the broader chronic kidney disease trial that included people without diabetes. Both are manufacturer-funded. This is the endpoint where the two are closest to interchangeable. **Magnitude:** [CREDENCE](https://pubmed.ncbi.nlm.nih.gov/30990260/) reported HR 0.70 (95% CI 0.59–0.82) for the composite kidney outcome with canagliflozin; [EMPA-KIDNEY](https://pubmed.ncbi.nlm.nih.gov/36331190/) reported HR 0.72 (0.64–0.82) with empagliflozin, at event rates of 13.1% versus 16.9%. #### Improved Blood Sugar Control Both lower glycated hemoglobin (HbA1c — the share of red blood cell protein coated in sugar, a three-month average of blood glucose) by removing glucose in the urine, independently of insulin, so the effect does not fade as beta-cell function declines. Canagliflozin at 300 mg adds a modest intestinal component. Randomized evidence and a large real-world comparison of closely matched patient groups both put the two agents at the same place; the drugs are separable from dapagliflozin but not from each other. **Magnitude:** Across 23 randomized trials, SGLT2 inhibitors added to metformin lowered HbA1c by 0.45–0.80 percentage points versus placebo ([Xu et al.](https://pubmed.ncbi.nlm.nih.gov/40010842/)). In matched real-world cohorts, six-month HbA1c was 7.65% on canagliflozin 300 mg versus 7.57% on empagliflozin 25 mg ([Blonde et al.](https://pubmed.ncbi.nlm.nih.gov/33180317/)). #### Body Weight Reduction Urinary loss of 60–100 g of glucose per day is a direct calorie drain, and both drugs produce weight loss that plateaus once compensatory eating catches up. The loss is roughly half fat mass and half fluid at first, with fat predominating later. Canagliflozin 300 mg ranks highest within the class on this endpoint in network meta-analysis, plausibly because of the added intestinal effect, but the absolute gap over empagliflozin is under one kilogram and no head-to-head trial exists. **Magnitude:** Class-wide, SGLT2 inhibitors reduced body weight by 0.88–2.67 kg versus placebo across 23 trials, with canagliflozin 300 mg ranked first ([Xu et al.](https://pubmed.ncbi.nlm.nih.gov/40010842/)); a 43-trial meta-analysis put the class mean at −1.88 kg (95% CI −2.11 to −1.66) ([Mazidi et al.](https://pubmed.ncbi.nlm.nih.gov/28546454/)). #### Lower Blood Pressure Both agents lower blood pressure through sodium loss and volume contraction rather than through any effect on vascular tone, which is why the reduction is achieved without a rise in heart rate. The magnitude is small but consistent across dozens of trials and applies to both drugs; it accounts for part, though not all, of the cardiovascular benefit. For someone already at target, the reduction is a reason to watch for lightheadedness rather than a benefit to seek. **Magnitude:** Meta-analysis of 43 randomized trials and 22,528 participants found a weighted mean reduction of 2.46 mmHg systolic (95% CI −2.86 to −2.06) and 1.46 mmHg diastolic (−1.82 to −1.09) ([Mazidi et al.](https://pubmed.ncbi.nlm.nih.gov/28546454/)). #### Reduced Cardiovascular and All-Cause Death ⚠️ Conflicted Empagliflozin's outcome trial showed a clear mortality reduction; canagliflozin's did not, and network ranking puts empagliflozin first for both death endpoints. Against that, formal indirect comparison finds no statistically significant difference between the agents, and one large routine-practice cohort found the opposite direction of effect on mortality for canagliflozin. All the trial evidence is manufacturer-funded. The net reading is that empagliflozin has the stronger direct mortality evidence, but the difference between the two agents has never been demonstrated in a comparison designed to detect it. **Magnitude:** [EMPA-REG OUTCOME](https://pubmed.ncbi.nlm.nih.gov/26378978/) reported cardiovascular death of 3.7% versus 5.9% (38% relative reduction) and death from any cause of 5.7% versus 8.3% (32% relative reduction); the [CANVAS Program](https://pubmed.ncbi.nlm.nih.gov/28605608/) reported no significant mortality benefit, and [Kani et al.](https://pubmed.ncbi.nlm.nih.gov/38293914/) found no significant between-agent difference across 21 trials. ### Medium 🟩 🟩 #### Lower Serum Urate and Fewer Gout Flares Glucose in the tubule competes with urate for reabsorption, so both drugs push urate out in the urine. In people with existing gout, starting an SGLT2 inhibitor rather than a comparator glucose-lowering drug reduced recurrent flares and gout-driven emergency visits. The controlled urate data are agent-specific for canagliflozin; the flare outcome data are observational and treat the class as a unit, so the two drugs cannot be ranked against each other here. **Magnitude:** Canagliflozin lowered serum urate by about 13% versus placebo at 26 weeks ([Davies et al.](https://pubmed.ncbi.nlm.nih.gov/25600248/)). In a matched cohort, gout flare rates were 52.4 versus 79.7 per 1,000 person-years, a rate ratio of 0.66 (95% CI 0.57–0.75) ([McCormick et al.](https://pubmed.ncbi.nlm.nih.gov/37487215/)). #### Reduced Risk of High Blood Potassium with Canagliflozin ⚠️ Conflicted In people with kidney disease taking blockers of the renin-angiotensin system (the hormone loop controlling salt and pressure), canagliflozin reduced episodes of hyperkalemia (dangerously high blood potassium) without causing low potassium. This matters because hyperkalemia is the usual reason those protective drugs get stopped. The conflict is that canagliflozin's own product labelling lists hyperkalemia as very common, up to 27%, in renally impaired populations. The net reading is that the randomized trial evidence is the stronger of the two and points to net protection. **Magnitude:** Investigator-reported hyperkalemia or potassium-binder initiation occurred at 32.7 versus 41.9 per 1,000 patient-years, HR 0.78 (95% CI 0.64–0.95) ([Neuen et al.](https://pubmed.ncbi.nlm.nih.gov/34423370/)). No matching randomized analysis exists for empagliflozin. #### Lower Recorded Dementia Incidence Pooled cohort data show fewer dementia diagnoses in people started on either drug than in those started on comparator glucose-lowering agents, with empagliflozin ranked most effective. Proposed mechanisms include improved brain insulin signalling and reduced small-vessel disease. The evidence is entirely observational — six cohort studies, no randomized outcome trial — so confounding by indication and by healthy-user effects cannot be excluded, and the between-agent ranking rests on indirect comparison. **Magnitude:** Across six cohorts and 845,433 people, HR for incident dementia was 0.69 (95% credible interval — the Bayesian counterpart of a confidence interval, 0.55–0.85) for empagliflozin and 0.75 (0.59–0.98) for canagliflozin; the two did not differ significantly from each other ([Meng et al.](https://pubmed.ncbi.nlm.nih.gov/42614235/)). ### Low 🟩 #### Increased Red Cell Mass Both drugs raise hemoglobin and hematocrit (the share of blood volume made up of red cells) within weeks, through restored oxygen sensing in the kidney cells that drive red cell production. The evidence is pooled placebo-controlled trials, which found no fall in anemia, so this stays a laboratory change. **Magnitude:** Across 17 placebo-controlled trials and 14,748 participants, hemoglobin rose by 5.60 g/L (95% CI 3.73–7.47) and hematocrit by 1.32 percentage points (1.21–1.44) ([Kanbay et al.](https://pubmed.ncbi.nlm.nih.gov/34273060/)). A network meta-analysis of 136 trials ranked canagliflozin highest for change from baseline in hemoglobin, with no reduction in anemia risk, odds ratio (OR — how much more likely an event is on the drug than on placebo) 0.93 (0.76–1.14) ([Sridharan & Sivaramakrishnan](https://pubmed.ncbi.nlm.nih.gov/42407295/)). #### Blunting of the Post-Meal Glucose Rise with Canagliflozin Because canagliflozin also inhibits the intestinal transporter, a 300 mg dose before a meal delays sugar absorption as well as spilling it in the urine. This is the one effect empagliflozin cannot reproduce. The evidence is a single crossover study in 20 healthy volunteers, not a clinical outcome. **Magnitude:** Canagliflozin 300 mg given 20 minutes before a mixed meal cut the post-meal glucose rise by 35% and the insulin rise by 43% over two hours, and slowed the appearance of oral glucose by 31% in the first hour ([Polidori et al.](https://pubmed.ncbi.nlm.nih.gov/23412078/)). ### Speculative 🟨 #### Lifespan Extension in Male Mice with Canagliflozin No human lifespan data exist for either drug. In genetically diverse mice, canagliflozin extended median male survival, an effect absent in females and never tested for empagliflozin. The basis is animal work only. #### Preserved Brain Insulin Sensitivity in Aged Mice with Canagliflozin In 30-month-old mice, canagliflozin improved insulin responsiveness in the hypothalamus and hippocampus and reduced age-related brain inflammation, mainly in males. No equivalent empagliflozin experiment or human counterpart exists. The basis is mechanistic animal data alone. ## Benefit-Modifying Factors - **Glucuronidation genotype:** Canagliflozin is cleared by UGT1A9 and UGT2B4 (liver enzymes that tag drugs with sugar for excretion). Reduced-function variants raise exposure and may amplify effect; empagliflozin uses a broader enzyme set and is less exposed to any single variant. - **SLC5A2 variants:** Common variants in SLC5A2 (the gene encoding the SGLT2 protein itself) alter baseline urinary glucose loss and are being studied as predictors of how much glucose lowering either drug delivers. No variant is yet actionable in practice. - **Baseline kidney filtration:** Glucose lowering from both drugs falls as the estimated glomerular filtration rate (eGFR — a calculated measure of kidney filtering capacity) declines, because less sugar is filtered. Heart and kidney protection, by contrast, persists to low filtration rates. - **Baseline biomarker levels:** Higher starting HbA1c, higher urine albumin, higher serum urate and higher blood pressure each predict a larger absolute change. Someone already metabolically optimized should expect small biomarker movement and correspondingly modest benefit. - **Sex:** Mouse lifespan benefit from canagliflozin was male-only, with higher drug blood levels in females. Human trial benefits show no consistent sex difference for either drug, so the animal asymmetry has no established human counterpart. - **Pre-existing heart or kidney disease:** Absolute benefit tracks baseline risk. In people with heart failure or albuminuric kidney disease the number needed to treat is small; in metabolically healthy people with normal kidneys, no outcome trial has ever measured the benefit. - **Age:** Older adults have higher event rates and so gain more in absolute terms from both drugs, but they also have the least fluid reserve. Trial populations averaged 63 to 66 years, so evidence in adults under 50 is thin. ## Potential Risks & Side Effects ### High 🟥 🟥 🟥 #### Genital Fungal Infections Sugar in the urine feeds yeast on the genital skin and mucosa, and this is the most common adverse effect of both drugs. Women are affected more often than men; uncircumcised men and anyone with a prior history are at higher risk. The evidence is consistent across dozens of randomized trials for both agents plus a controlled cohort using genital infection as a positive control. It is usually treatable with standard antifungal therapy and rarely forces discontinuation. **Magnitude:** Canagliflozin labelling reports female genital fungal infection in up to 11.4% and recurrent male genital fungal infection in 22%; empagliflozin's outcome trial reported genital infection in 6.4% versus 1.8% on placebo ([Zinman et al.](https://pubmed.ncbi.nlm.nih.gov/26378978/)). A matched cohort found HR 2.15 (95% CI 1.39–3.30) ([McCormick et al.](https://pubmed.ncbi.nlm.nih.gov/37487215/)). Network meta-analysis finds both agents significantly above placebo ([Li et al.](https://pubmed.ncbi.nlm.nih.gov/37701900/)). #### Increased Urination and Volume Depletion The osmotic diuresis that drives the benefit also drives the nuisance: more urine, more night waking, more thirst, and in susceptible people postural dizziness or fainting. Risk concentrates in those on loop diuretics, those over 75, and those with low starting blood pressure. Network meta-analysis of 113 trials finds a significant excess of urinary frequency for empagliflozin, and canagliflozin labelling records volume-related events as common. Most cases resolve with fluid adjustment or diuretic dose reduction. **Magnitude:** Urinary frequency odds ratio 5.81 (95% CI 1.79–32.97) for empagliflozin versus placebo across 113 randomized trials ([Li et al.](https://pubmed.ncbi.nlm.nih.gov/37701900/)); canagliflozin labelling lists postural hypotension (a blood pressure drop on standing), dehydration and syncope (fainting) at 1–10%. #### Early Fall in Estimated Kidney Filtration Rate Both drugs cause an immediate, reversible drop in measured filtration in the first weeks, because they narrow the vessel feeding the filter. This is the mechanism of the long-term benefit, not evidence of harm, and the trajectory recovers and then outperforms placebo. The practical hazard is misreading it: the drug gets stopped, or a clinician adds contrast or a non-steroidal anti-inflammatory on top of a volume-depleted patient and precipitates genuine acute kidney injury. **Magnitude:** Both pivotal kidney trials show an initial eGFR decline of roughly 3–5 mL/min/1.73 m² followed by a slower long-term slope than placebo, with the composite kidney outcome improving at HR 0.70 for canagliflozin ([Perkovic et al.](https://pubmed.ncbi.nlm.nih.gov/30990260/)) and HR 0.72 for empagliflozin ([Herrington et al.](https://pubmed.ncbi.nlm.nih.gov/36331190/)). #### Small Adverse Shift in Blood Lipids Both drugs nudge cholesterol upward, an effect usually attributed to reduced clearance of lipoprotein particles during the fuel shift toward fat. Canagliflozin labelling lists raised low-density lipoprotein cholesterol (LDL — the cholesterol-carrying particle most strongly tied to arterial disease) as a common finding. The evidence is a pooled analysis of 43 randomized trials. The measured magnitude is far too small to offset the cardiovascular benefit, and triglycerides move the other way. **Magnitude:** Across 43 trials, total cholesterol rose by 0.77 mg/dL (95% CI 0.33–1.21) while triglycerides fell by 2.08 mg/dL (−2.51 to −1.64) ([Mazidi et al.](https://pubmed.ncbi.nlm.nih.gov/28546454/)); canagliflozin labelling places raised LDL cholesterol in the 1–10% band. ### Medium 🟥 🟥 #### Ketoacidosis at Normal or Near-Normal Blood Glucose Both drugs shift metabolism toward ketones and lower insulin, so acidosis can develop while blood glucose still looks acceptable — which delays recognition. Triggers are fasting, very low carbohydrate intake, acute illness, alcohol and surgery. The signal comes from a 417,514-person multi-database cohort and from post-marketing reports; randomized trials are underpowered for it and show no difference from placebo. For this audience the risk is elevated by the same practices they tend to adopt. **Magnitude:** Versus dipeptidyl peptidase-4 inhibitors (another class of oral diabetes drugs, which work by prolonging the body's own insulin-release signal), HR 3.58 (95% CI 2.13–6.03) for canagliflozin and 2.52 (1.23–5.14) for empagliflozin, on an absolute incidence of 2.03 versus 0.75 per 1,000 person-years ([Douros et al.](https://pubmed.ncbi.nlm.nih.gov/32716707/)). #### Low Blood Sugar When Combined with Insulin or Sulfonylureas Neither drug causes hypoglycemia alone, because its action does not depend on insulin. Combined with insulin or an insulin-releasing sulfonylurea, however, the added glucose loss makes existing doses excessive. This is the single most predictable adverse interaction of either agent and is fully avoidable by pre-emptive dose reduction of the partner drug. Network meta-analysis finds no excess of severe hypoglycemia for either agent against placebo when used alone. **Magnitude:** Canagliflozin labelling reports hypoglycemia in up to 49% when combined with insulin or an insulin secretagogue (a drug that pushes the pancreas to release more insulin), against no significant excess for either agent as monotherapy versus placebo across 113 trials ([Li et al.](https://pubmed.ncbi.nlm.nih.gov/37701900/)). ### Low 🟥 #### Lower-Limb Amputation with Canagliflozin ⚠️ Conflicted Canagliflozin's cardiovascular program found amputations, mostly toe or midfoot, at nearly twice the placebo rate, producing a 2017 boxed warning. Its kidney trial found no imbalance and regulators withdrew the warning in 2020. No mechanism was established. The net reading: real in one dataset, absent in others. **Magnitude:** The [CANVAS Program](https://pubmed.ncbi.nlm.nih.gov/28605608/) reported 6.3 versus 3.4 amputations per 1,000 patient-years, HR 1.97 (95% CI 1.41–2.75). Pooled canagliflozin trials give relative risk 1.59 (1.26–2.01) under fixed effects (a pooling method that assumes a single true effect across trials), while empagliflozin trials show no excess ([Heyward et al.](https://pubmed.ncbi.nlm.nih.gov/32502190/)). #### Bone Fracture and Hip Bone Density Loss with Canagliflozin ⚠️ Conflicted Canagliflozin raised bone turnover markers and reduced hip bone density, and fractures clustered in the same program that produced the amputation signal, appearing as early as 12 weeks. Other trial pools showed no imbalance. The net reading is a real but small skeletal effect confined to canagliflozin. **Magnitude:** Total hip bone mineral density fell up to 1.2% over two years on canagliflozin, with no change at other skeletal sites ([Alba et al.](https://pubmed.ncbi.nlm.nih.gov/27046479/)); across 117 trials neither agent significantly changed fracture risk ([Zhang et al.](https://pubmed.ncbi.nlm.nih.gov/34721294/)). #### Urinary Tract Infection ⚠️ Conflicted Sugar-rich urine is a plausible growth medium, and both labels warn of pyelonephritis (kidney infection) and urosepsis (bloodstream infection from the urinary tract). Controlled trials show no excess for either agent. The net reading is a genuine but rare post-marketing event, not a measurable trial-level risk. **Magnitude:** Across 113 randomized trials, neither empagliflozin nor canagliflozin showed a significant increase in urinary tract infection versus placebo; only dapagliflozin (odds ratio 1.33, 95% CI 1.10–1.62) and remogliflozin did ([Li et al.](https://pubmed.ncbi.nlm.nih.gov/37701900/)). #### Fournier Gangrene A rare, rapidly destructive infection of the tissue between the anus and genitals, requiring emergency surgery. Regulators identified 55 cases across this drug class between 2013 and 2019, in both sexes. A meta-analysis of 84 trials found no measurable excess, but events were too few to exclude one. **Magnitude:** Odds ratio 0.41 (95% CI 0.09–1.82) across 84 trials, 42,415 exposed participants — an interval too wide to exclude harm ([Silverii et al.](https://pubmed.ncbi.nlm.nih.gov/31637829/)); the regulator's case series reported 55 events, all requiring surgical debridement, with three deaths ([Bersoff-Matcha et al.](https://pubmed.ncbi.nlm.nih.gov/31060053/)). #### Bladder Cancer Signal with Empagliflozin ⚠️ Conflicted A meta-analysis of 46 trials found no overall cancer excess but a bladder cancer signal driven by empagliflozin, alongside an apparently protective gastrointestinal finding for canagliflozin. Both rest on tiny event counts in trials far shorter than cancer latency. The net reading: chance findings until longer data arrive. **Magnitude:** Bladder cancer odds ratio 3.87 (95% CI 1.48–10.08) for the class and 4.49 (1.21–16.73) for empagliflozin; gastrointestinal cancer odds ratio 0.15 (0.04–0.60) for canagliflozin; overall cancer 1.14 (0.96–1.36) ([Tang et al.](https://pubmed.ncbi.nlm.nih.gov/28725912/)). #### Higher Event Rates with Canagliflozin in Routine-Practice Data ⚠️ Conflicted A closely matched cohort of 24,078 adults without prior heart disease found more major cardiac events and more deaths on canagliflozin than on selective SGLT2 inhibitors, contradicting the randomized evidence. Residual confounding is the obvious explanation. The net reading is that randomized data govern here. **Magnitude:** Versus other SGLT2 inhibitors, canagliflozin carried HR 1.23 (95% CI 1.14–1.33) for the composite of myocardial infarction (heart attack), stroke or death, HR 1.49 (1.33–1.68) for death from any cause, and HR 1.35 (1.02–1.79) for hemorrhagic stroke ([Kornelius et al.](https://pubmed.ncbi.nlm.nih.gov/41190032/)). ### Speculative 🟨 #### Increased Intestinal Tumor Burden with Canagliflozin in Animals In a tumor-prone mouse strain, canagliflozin increased the gut area involved by adenomas (benign growths that can turn cancerous), mainly in females. The proposed explanation is redistributed glucose absorption. The basis is animal data alone. ## Risk-Modifying Factors - **Glucuronidation genotype:** Reduced-function UGT1A9 or UGT2B4 variants raise canagliflozin exposure, plausibly amplifying dose-related effects such as volume depletion. Empagliflozin's clearance across a wider enzyme set makes it the more predictable choice for anyone with known variants. - **Baseline ketones and insulin reserve:** Low fasting insulin, low C-peptide (a marker of how much insulin the pancreas still makes) or a baseline ketone level already above 0.6 mmol/L raise ketoacidosis risk sharply, particularly on carbohydrate-restricted diets. - **Baseline kidney filtration and blood pressure:** eGFR under 30 mL/min/1.73 m², systolic pressure under 110 mmHg, or established diuretic use each raise the chance of symptomatic volume depletion and acute kidney injury after starting either drug. - **Sex:** Women carry the higher genital fungal infection rate; uncircumcised men carry the higher rate of balanitis (inflammation of the penis head). In mice, female canagliflozin blood levels ran above male levels, and late-life treatment shortened female survival. - **Pre-existing conditions:** Peripheral arterial disease, neuropathy, prior amputation or active foot ulcer concentrate the canagliflozin amputation risk. Osteoporosis concentrates the fracture risk. Type 1 diabetes and recurrent ketoacidosis concentrate the acidosis risk. - **Age:** Adults over 75 have less fluid reserve, more falls and more fractures, and are more likely to be on diuretics and renin-angiotensin blockers, compounding the volume effects of either drug. ## Key Interactions & Contraindications - **Insulin and sulfonylureas (glimepiride, glipizide, glyburide):** Caution, not contraindication. Additive glucose lowering produces hypoglycemia. Mitigation is to cut the insulin dose by 10–20% and halve the sulfonylurea at initiation, then retitrate. - **Loop and thiazide diuretics (furosemide, torsemide, hydrochlorothiazide):** Caution. Additive volume depletion causes orthostatic hypotension (a blood pressure drop on standing), falls and acute kidney injury. Mitigation is a 25–50% diuretic dose reduction on the day of initiation. - **UGT enzyme inducers (rifampin, phenytoin, phenobarbital, ritonavir):** Monitor. These roughly halve canagliflozin exposure and blunt its effect; the label permits increasing to 300 mg daily. Empagliflozin is less affected and is the simpler co-prescription. - **Digoxin:** Monitor. Canagliflozin 300 mg raises digoxin exposure by about 20% through P-glycoprotein (a cellular drug pump), risking digoxin toxicity. Mitigation is a digoxin level after two weeks of combined use. - **Renin-angiotensin blockers and mineralocorticoid antagonists (the latter block a salt-retaining hormone; lisinopril, losartan, spironolactone, finerenone):** Caution. Additive blood pressure lowering, but favorable potassium effects. Deliberately combined in kidney protection; check potassium and creatinine at four weeks. - **Over-the-counter non-steroidal anti-inflammatories (ibuprofen, naproxen, diclofenac gel):** Caution. Combined with volume depletion they precipitate acute kidney injury. Mitigation is to avoid regular use, and to stop them entirely during any illness with vomiting or diarrhoea. - **Over-the-counter decongestants and antihistamine-decongestant combinations (pseudoephedrine, phenylephrine):** Monitor. These raise blood pressure and reduce urinary flow, partially opposing the drug and worsening urinary retention symptoms in men with prostate enlargement. - **Glucose-lowering supplements (berberine, chromium picolinate, alpha-lipoic acid, cinnamon extract, gymnema):** Caution. These have additive glucose-lowering effects and can push readings low when stacked on either drug plus insulin. Mitigation is to separate their introduction by four weeks. - **Diuretic and potassium-losing supplements (dandelion extract, juniper, high-dose caffeine):** Caution. Additive fluid and electrolyte loss on top of osmotic diuresis. Mitigation is deliberate sodium and fluid replacement, and a potassium check at four weeks. - **Ketogenic diets, prolonged fasting and endurance events:** Caution. All three deepen the ketone shift both drugs create and are the commonest precipitants of ketoacidosis at normal glucose. Mitigation is a carbohydrate floor and home ketone testing. **Populations who should avoid Empagliflozin vs. Canagliflozin:** - Type 1 diabetes, or type 2 diabetes with C-peptide below 0.6 ng/mL, because of the ketoacidosis rate - Any prior episode of ketoacidosis on this drug class - eGFR below 20 mL/min/1.73 m² for new initiation, and dialysis at any filtration rate - Prior lower-limb amputation, active foot ulcer or critical limb ischemia (severely reduced blood flow to the leg) — canagliflozin specifically - Established osteoporosis with a T-score at or below −2.5 — canagliflozin specifically - Pregnancy in the second and third trimesters, and breastfeeding - Recurrent Fournier gangrene, necrotizing perineal infection or recurrent urosepsis - Severe hepatic impairment, Child-Pugh Class C — canagliflozin is not recommended - Documented hypersensitivity, angioedema (rapid deep swelling of the face, lips or throat) or anaphylaxis to either agent ## Risk Mitigation Strategies - **Low starting dose held for 4–12 weeks:** Beginning at empagliflozin 10 mg or canagliflozin 100 mg daily for 4–12 weeks limits volume depletion, dizziness and falls, and most cardiorenal benefit is already present at the lower dose. - **Pre-emptive diuretic reduction:** Cutting any loop or thiazide diuretic by 25–50% on the day of initiation prevents the orthostatic hypotension and acute kidney injury that follow additive volume loss, with the dose restored later if fluid overload returns. - **Carbohydrate floor of 100 g daily:** Maintaining at least 100 g of carbohydrate per day and avoiding ketogenic protocols entirely prevents the ketoacidosis that both drugs cause at normal blood glucose. This is the single highest-yield precaution for this audience. - **Sick-day and surgery rules:** Stopping the drug at the onset of any illness with vomiting, diarrhoea or reduced intake, and at least 3 days before elective surgery, prevents both ketoacidosis and volume-depletion kidney injury. - **Home ketone meter:** Keeping a blood beta-hydroxybutyrate meter and testing whenever nausea, breathlessness or unusual fatigue occurs catches ketoacidosis while blood glucose still reads normal, which is when it is otherwise missed. - **Genital hygiene protocol:** Daily washing and thorough drying after urination, prompt treatment of the first yeast episode, and a standby antifungal prescription reduce recurrent genital fungal infection and the rare progression to perineal necrotizing infection. - **Daily foot inspection on canagliflozin:** Checking feet daily for sores, blisters and colour change, and stopping the drug at the first non-healing lesion, addresses the amputation signal confined to canagliflozin's cardiovascular program. - **Baseline and biennial bone density scan on canagliflozin:** A bone density scan before starting and every two years thereafter, with a switch to empagliflozin if the T-score falls below −2.0, addresses canagliflozin's documented hip bone density loss. - **Agent selection where foot or bone risk is present:** Selecting empagliflozin where peripheral arterial disease, neuropathy or osteoporosis already exists avoids both canagliflozin-specific signals at no measurable cost in cardiorenal or glycemic benefit. - **Sodium and fluid replacement:** Adding 1–2 g of sodium daily and 500–750 mL of extra fluid during the first month offsets osmotic diuresis and prevents the lightheadedness that otherwise causes people to stop the drug prematurely. ## Therapeutic Protocol - **Standard empagliflozin dosing:** 10 mg once daily, increased to 25 mg after 4–12 weeks if glucose lowering is the goal. Cardiorenal indications use 10 mg only; no outcome trial has shown 25 mg to be superior for those endpoints. - **Standard canagliflozin dosing:** 100 mg once daily, increased to 300 mg after 4–12 weeks. The 300 mg dose is where the intestinal transporter effect appears; 100 mg is the dose used in the kidney outcome trial. - **Cardiorenal-first approach:** Practitioners prioritising heart and kidney endpoints favor empagliflozin at 10 mg, citing its mortality trial and its cleaner limb and bone record. Diabetes and kidney professional bodies endorse either agent for these indications. - **Organizational interest note:** The diabetes and kidney professional bodies whose endorsement is cited above are physician-membership organizations. Their members earn no direct revenue from which agent is chosen, but their guideline panels routinely include authors with payments from both manufacturers. - **Glycemic and weight-first approach:** Practitioners prioritising glucose peaks and weight favor canagliflozin 300 mg for its added intestinal action. [Peter Attia](https://peterattiamd.com/ama53/) has discussed this class's use as a possible aging-slowing agent rather than only a glucose drug. - **Timing of dose:** Both are taken once daily in the morning before the first meal. Canagliflozin's pre-meal timing is functional rather than conventional: the intestinal effect requires drug in the gut lumen when food arrives. - **Half-life and dosing frequency:** Empagliflozin's half-life is about 12 hours, canagliflozin's 10.6 hours at 100 mg and 13.1 hours at 300 mg. Both reach steady state within four days, and both are single daily doses; splitting confers no advantage. - **Genetic influences on dose choice:** Reduced-function UGT1A9 or UGT2B4 variants raise canagliflozin exposure and argue for staying at 100 mg. No pharmacogenetic test is validated for either drug, so genotype currently informs caution rather than dosing. - **Sex-based differences:** Human trials show no sex difference in glucose lowering or cardiorenal benefit for either drug, and no sex-specific dosing exists. Women should expect a higher genital infection rate at any dose. - **Age-related adjustments:** No dose reduction is required by age alone, but over 75 the practical approach is empagliflozin 10 mg with a diuretic reduction, since fall and fracture consequences outweigh the marginal glycemic gain from higher doses. - **Baseline biomarkers guiding response:** Higher starting HbA1c and higher urine albumin predict larger movement. Below an eGFR of 45 mL/min/1.73 m², glucose lowering largely disappears while cardiorenal protection persists, so glycemic dose escalation stops making sense. - **Pre-existing conditions guiding choice:** Heart failure, albuminuric kidney disease and gout favor either agent. Peripheral arterial disease and osteoporosis favor empagliflozin. Obesity with large post-meal glucose excursions is the one profile favoring canagliflozin 300 mg. ## Discontinuation & Cycling - **Intended duration:** Both are intended as indefinite therapy. Every outcome trial ran continuous treatment for two to five years, and no trial has tested a defined stopping point or a fixed course for either drug. - **Loss of effect on withdrawal:** Blinded withdrawal after years of empagliflozin raised the rate of cardiovascular death or heart failure hospitalization within roughly 30 days, with weight, glucose, blood pressure, urate and filtration rate all rebounding toward baseline. - **Withdrawal effects:** There is no dependence or abstinence syndrome. What returns is the untreated disease state — fluid retention, higher glucose, higher urate — plus a transient rise in measured filtration rate that can be mistaken for improvement. - **Tapering protocol:** No pharmacological taper is needed, since neither drug shows rebound receptor effects. Where the drug was combined with a reduced diuretic dose, the diuretic must be restored as the SGLT2 inhibitor stops, or fluid overload follows. - **Temporary interruption:** Planned holds for illness, surgery or endurance events run the shortest interval consistent with safety — typically 3 days before a procedure and until oral intake is normal afterwards, with the previous dose then resumed. - **Cycling:** No evidence supports cycling either drug. The proposal that intermittent dosing might preserve a ketone benefit while limiting genital infections has never been tested, and every documented benefit accrued under continuous dosing. ## Sourcing and Quality - **Prescription status:** Both are prescription-only in every major market. There is no legitimate supplement, compounded or over-the-counter route, and compounding pharmacies have no role for either small-molecule tablet. - **Brand supply:** Empagliflozin is supplied as Jardiance by Boehringer Ingelheim with Eli Lilly; canagliflozin as Invokana by Janssen. Both are manufactured to full pharmacopoeial specification, so purity and content uniformity are not the practical concern here. - **Generic availability:** A generic canagliflozin was approved for Apotex in April 2026 but is not yet widely marketed. No United States generic empagliflozin is expected before roughly 2029. Availability, not quality, is the current constraint. - **What to look for:** An approved generic carries an abbreviated new drug application number and a bioequivalence rating; those are the markers worth checking. Any source selling either drug without a prescription is outside the regulated supply chain. - **Pharmacy verification:** Legitimate online pharmacies carry National Association of Boards of Pharmacy accreditation or a .pharmacy domain. Unaccredited overseas sellers of these two drugs are a documented source of counterfeit and under-dosed tablets. - **Formulation considerations:** Both come only as immediate-release tablets, and fixed-dose combinations with metformin or linagliptin exist for both. Combination products complicate the sick-day hold, since the partner drug must be stopped too. ## Practical Considerations - **Time to effect:** Urinary glucose loss begins with the first dose. Blood pressure and glucose respond within 1–2 weeks, weight plateaus at 12–26 weeks, heart failure benefit appears within weeks, and the kidney slope benefit takes 1–2 years to separate from placebo. - **Common pitfall — stopping at the filtration dip:** The expected early fall in estimated filtration rate is routinely misread as kidney damage, and the drug is stopped at exactly the point where the long-term benefit is being set up. Rechecking at four weeks resolves it. - **Common pitfall — stacking with a ketogenic diet:** Combining either drug with carbohydrate restriction or extended fasting is the commonest route to ketoacidosis in this population, and blood glucose gives no warning because it stays normal. - **Common pitfall — not holding before surgery:** Failure to stop 3 days before an elective procedure is a leading cause of post-operative ketoacidosis, which is frequently missed because glucose readings look unremarkable. - **Regulatory status:** Both are fully approved for type 2 diabetes; empagliflozin additionally for heart failure and chronic kidney disease, canagliflozin for diabetic kidney disease. Use in healthy people for longevity is off-label and unstudied for that indication. - **Cost and payer incentives:** Both list near $600–700 monthly in the United States. Once generic canagliflozin reaches supply, institutional payers gain a systematic incentive to favor it, a potential source of structural bias in formulary policy, guideline formation and research funding. ## Interaction with Foundational Habits - **Sleep:** Direct and mostly negative. Osmotic diuresis increases night waking to urinate, especially in the first month and at higher doses, and this is the commonest reason people abandon either drug. Practical response is to take the dose on waking, front-load fluids before mid-afternoon, and expect the effect to settle by week four. - **Nutrition:** Direct and bidirectional. Both drugs deepen the ketone shift, so carbohydrate restriction and extended fasting sharply raise ketoacidosis risk; a floor of about 100 g of carbohydrate daily is protective. Canagliflozin 300 mg taken before the largest meal exploits its intestinal effect. Osmotic diuresis also raises sodium and fluid requirements. - **Exercise:** Indirect and mildly blunting. Reduced circulating glucose and lower plasma volume can cut endurance performance and raise heat strain, and long events combined with low intake are a documented ketoacidosis trigger. Practical response is to hold the dose before endurance events, hydrate with electrolytes, and keep protein intake up to protect lean mass. - **Stress management:** Indirect and potentiating. Stress hormones raise glucose and ketone production, so acute physical or psychological stress on top of either drug widens the window for ketoacidosis at normal glucose. No effect of either drug on cortisol has been demonstrated. Practical response is to test ketones during illness or acute stress. ## Monitoring Protocol & Defining Success Before starting either drug, a baseline set establishes both the safety floor and the yardstick against which any effect will be judged: kidney filtration and urine albumin, a full electrolyte panel with potassium and magnesium, serum urate, a lipid panel, HbA1c and fasting glucose, hematocrit, seated and standing blood pressure, body weight and composition, and a bone density scan where canagliflozin and any fracture risk factor coincide. Ongoing testing follows a fixed cadence: electrolytes, filtration rate and standing blood pressure at 4 weeks; the full baseline panel repeated at 3 months and 6 months; then every 6–12 months indefinitely, with an unscheduled panel after any illness, procedure or dose change. | Biomarker | Optimal Functional Range | Why Measure It? | Context/Notes | | --- | --- | --- | --- | | HbA1c | 5.0–5.6% if not diabetic; below 6.5% if diabetic | Tracks the core glucose effect of either drug | No fasting needed. Both agents raise red cell turnover slightly, which biases the reading marginally low. | | Fasting glucose | 75–90 mg/dL (4.2–5.0 mmol/L) | Detects response and the low-glucose risk from insulin or sulfonylurea combinations | Requires a 10–12 hour fast. Conventional labs call anything under 100 mg/dL normal; functional targets are tighter. | | eGFR | Above 90 mL/min/1.73 m², or a stable slope after the first month | Distinguishes the expected early dip from genuine kidney injury | Conventional labs flag only values below 60 mL/min/1.73 m²; the functional target sits well above that. A 3–5 mL/min drop in the first 4 weeks reflects changed blood flow inside the filter and is expected. A recheck at 4 weeks, not sooner, is the informative timepoint. | | Urine albumin-to-creatinine ratio | Below 10 mg/g | The most responsive marker of kidney benefit from either drug | First-morning sample. Conventional labs flag only above 30 mg/g; the functional target is well below that. | | Serum potassium | 4.0–4.5 mmol/L | Both drugs interact with potassium handling, especially alongside kidney-protective blood pressure drugs | Fist clenching during the draw falsely raises the value. Conventional range extends to 5.2 mmol/L. | | Serum magnesium | 2.0–2.4 mg/dL | Both agents raise magnesium; a fall instead suggests excess diuresis | Conventional ranges start around 1.7 mg/dL, below the functional floor. Best paired with potassium. Red cell magnesium is the more sensitive test where the serum value looks normal. | | Serum urate | 3.5–6.0 mg/dL in men, 3.0–5.5 mg/dL in women | Confirms the urate-lowering effect and predicts gout flare protection | Requires an 8-hour fast and 24 hours without alcohol. Conventional upper limits run to 7.0–8.0 mg/dL. | | Hematocrit | 40–48% in men, 36–44% in women | Both drugs raise red cell mass; an excessive rise signals dehydration or over-diuresis | Drawn seated after 5 minutes' rest. A rise above 52% warrants fluid review before dose escalation. | | LDL cholesterol | Below 100 mg/dL, or lower according to individual cardiovascular risk | Both drugs nudge cholesterol upward, and the shift is worth quantifying rather than assuming | Direct measurement preferred over calculation. A repeat at 3 months captures the shift, which appears early and then stabilizes. | | Beta-hydroxybutyrate (blood ketone) | Below 0.6 mmol/L on a normal carbohydrate intake | The only marker that catches ketoacidosis while blood glucose still reads normal | Home fingerstick meter. Testing is symptom-triggered — nausea, breathlessness or unusual fatigue — rather than scheduled. | | Blood pressure, seated and standing | Below 120/80 mmHg seated, with under 20 mmHg systolic drop on standing | Captures both the intended blood pressure benefit and the volume depletion risk | The standing reading is taken after 3 minutes and is the value that predicts falls. | | Bone mineral density by DXA | T-score at or above −1.0 | Canagliflozin reduces hip bone density; this is the only way to see it | DXA is dual-energy X-ray absorptiometry, the standard bone density scan. Baseline then every 2 years on canagliflozin, hip and lumbar spine sites. | | Lean body mass by DXA | No established target for this drug class; track change from the individual's own baseline | Separates fat loss from lean loss during the weight reduction both drugs produce | The same scanner and the same hydration state each time; otherwise the comparison is meaningless. | Qualitative markers worth tracking alongside the laboratory panel: - Night-time urination frequency, which peaks in the first month and should settle thereafter - Lightheadedness or unsteadiness on standing, particularly on rising in the morning - Genital itching, discharge or discomfort, which should trigger treatment at first appearance - Perineal pain, swelling or redness with fever, which is an emergency rather than a symptom to monitor - Exercise tolerance and perceived effort during endurance work, which may fall in the first weeks - Thirst and daily fluid intake, as a proxy for whether replacement is keeping pace with diuresis - Foot sensation, skin integrity and healing of minor lesions, especially on canagliflozin ## Emerging Research - **Direct head-to-head comparison:** [NCT05220917](https://clinicaltrials.gov/study/NCT05220917) compares four second-line strategies in type 2 diabetes across 781,430 people, with cardiovascular and kidney endpoints. It is the largest effort to date to separate agents within this class rather than against placebo. - **Combination with an aldosterone synthase inhibitor:** [NCT06531824](https://clinicaltrials.gov/study/NCT06531824) tests vicadrostat plus empagliflozin against empagliflozin alone in 11,000 people with chronic kidney disease, a phase 3 trial sponsored by Boehringer Ingelheim, the manufacturer of empagliflozin. - **Sex-specific effects:** [NCT07188545](https://clinicaltrials.gov/study/NCT07188545) examines sex-based heterogeneity of this drug class on cardiovascular, kidney and metabolic outcomes across an estimated 23.3 million real-world records, with major adverse cardiovascular events as the primary endpoint. It bears directly on whether the male-only mouse lifespan finding has a human counterpart. - **Cognitive endpoints for canagliflozin:** [NCT07711171](https://clinicaltrials.gov/study/NCT07711171) is a phase 4 trial of canagliflozin in 60 people with type 2 diabetes and mild cognitive impairment, the first attempt to test the mouse brain-aging findings in people. - **Pharmacogenetics of response:** [NCT02891954](https://clinicaltrials.gov/study/NCT02891954) enrols 700 people to map genetic determinants of canagliflozin response. Its output would determine whether the glucuronidation variants discussed above become actionable rather than merely cautionary. - **Mechanism that could strengthen the case:** Whether the senescent-cell-clearing account holds in humans is the pivotal open question, since [Katsuumi et al., 2024](https://pubmed.ncbi.nlm.nih.gov/38816549/) showed canagliflozin clearing worn-out cells and extending life in fast-aging mice, and no human trial has yet measured that cell burden on either drug. - **Evidence that could weaken the case:** [Butler et al., 2024](https://pubmed.ncbi.nlm.nih.gov/38587237/) found empagliflozin did not reduce the primary composite endpoint after myocardial infarction, and [Kornelius et al., 2025](https://pubmed.ncbi.nlm.nih.gov/41190032/) reported worse routine-practice outcomes for canagliflozin. Both cut against a uniform class benefit. ## Conclusion Empagliflozin and canagliflozin do the same core thing: they make the kidneys shed sugar, and in doing so they lower weight and blood pressure, reduce hospital admissions for heart failure, and slow the loss of kidney function. On those measures the evidence is strong, replicated, and does not meaningfully separate the two drugs. Uric acid falls as well, on thinner evidence. Where they part company is at the edges. Empagliflozin has the clearer evidence for lower death rates and a quieter record on limbs and bones. Canagliflozin has an extra action in the gut that blunts sugar peaks after meals, and it is the only one of the two ever shown to lengthen life in animals — a male-only finding, in mice, with no human counterpart. The costs are shared and mostly manageable: fungal infections of the genital skin, more urination, and a form of acid build-up in the blood that can happen while sugar readings still look normal, which is the risk that matters most for anyone restricting carbohydrate or fasting. The evidence base has two structural weaknesses worth naming. Almost every pivotal trial was paid for by the company selling the drug, and the professional bodies whose recommendations shape practice draw no direct revenue from the choice but seat authors funded by both makers. The two drugs have never been compared head to head in a trial designed to tell them apart. **[Top](#top) - [Benefits](#expected-benefits) - [Risks](#potential-risks--side-effects) - [Protocol](#therapeutic-protocol)**