Pioglitazone, sold as Actos, is an oral diabetes drug from the thiazolidinedione (TZD) class. It works by making muscle, fat and liver respond better to the body’s own insulin. The FDA first approved it in 1999 for adults with type 2 diabetes, and it is also widely studied in fatty liver disease (MASH, formerly called NASH) and in insulin-resistant people after a stroke.
The single most important fact is its boxed warning: thiazolidinediones, including pioglitazone, can cause or worsen congestive heart failure. The label also warns about bladder cancer, dose-related edema (fluid swelling), fractures in women and weight gain (FDA label).
This page summarizes what the FDA label, the large outcome trials and liver-disease studies report. It is for education only and is not medical advice.
| Property | Detail |
|---|---|
| Drug class | Thiazolidinedione; PPAR-gamma agonist (insulin sensitizer) |
| Brand and approval | Actos (Takeda); initial US approval 1999; generics and combinations with metformin are listed on DailyMed |
| Approved use | Adjunct to diet and exercise to improve blood sugar control in adults with type 2 diabetes |
| Route and strengths | Oral tablets, 15 mg, 30 mg and 45 mg |
| Labeled dose | 15 or 30 mg once daily; up to 45 mg once daily (15 mg max with gemfibrozil) |
| Half-life (label) | Pioglitazone 3–7 h; active metabolites M-III and M-IV 16–24 h |
| Boxed warning | Congestive heart failure; starting it in NYHA class III or IV heart failure is contraindicated |
| US status | Prescription only (Rx) |
| WADA status | Not named on the 2026 Prohibited List (S4.4 names PPAR-delta agonists, a different receptor) |
What it is
Pioglitazone is a small synthetic molecule built around a thiazolidinedione ring. The hydrochloride salt has the formula C19H20N2O3S·HCl and a molecular weight of 392.90. It is made and used as a racemic mixture; the two mirror-image forms convert into each other in the body and have the same activity (FDA label).
Mechanism: a PPAR-gamma insulin sensitizer
Pioglitazone activates peroxisome proliferator-activated receptor gamma (PPAR-gamma), a nuclear receptor found in fat tissue, skeletal muscle and liver. When activated, it changes the transcription of insulin-responsive genes involved in glucose and fat metabolism. The label describes the result as lower insulin resistance in the periphery and the liver, more insulin-dependent glucose uptake, and less glucose output from the liver (FDA label).
Two practical points follow from this:
- It needs the body’s own insulin. Pioglitazone does not make the pancreas release insulin (it is not an insulin secretagogue). It does not lower blood sugar in animal models that lack insulin, and the label says it should not be used for type 1 diabetes or diabetic ketoacidosis.
- It lowers insulin levels as well as glucose. In type 2 diabetes, the label reports lower plasma glucose, lower plasma insulin and lower HbA1c (average blood sugar over about 3 months).
Where body fat goes
Pioglitazone usually increases total body fat while improving insulin sensitivity. In a small 16-week study of 13 people with type 2 diabetes taking 45 mg/day, MRI showed that fat moved from the belly’s internal (visceral) depot to the fat under the skin (Miyazaki et al., JCEM 2002):
| Measure (mean) | Before | After 16 weeks |
|---|---|---|
| Body weight | 82 kg | 85 kg |
| Fat mass | 27 kg | 30 kg |
| Subcutaneous fat area at L4–5 | 301 cm² | 342 cm² |
| Visceral fat area at L4–5 | 144 cm² | 131 cm² |
| HbA1c | 8.6% | 7.2% |
This was an uncontrolled study with no placebo group, so it shows the direction of change rather than a precise effect size.
Pharmacokinetics
From the Actos label (FDA label):
| Parameter | Value |
|---|---|
| Time to peak | Within 2 h; food delays this to 3–4 h without changing total absorption |
| Protein binding | Over 99%, mainly to albumin |
| Metabolism | Mainly CYP2C8, less CYP3A4; active metabolites M-III (keto) and M-IV (hydroxyl) |
| Steady state | Within 7 days of once-daily dosing |
| Half-life | Pioglitazone 3–7 h; M-III and M-IV 16–24 h |
| Elderly | Half-life ≈10 h vs ≈7 h in younger adults; AUC ≈21% higher |
| Women vs men | Peak level and AUC 20–60% higher in women |
| Elimination | 15–30% of the dose recovered in urine as metabolites; most presumed excreted in bile |
| Kidney impairment | No change in half-life; no dose adjustment |
Because the active metabolites reach levels equal to or above the parent drug and last much longer, the parent drug’s short half-life understates how long the effect lasts. Pioglitazone itself makes up only 20–25% of total drug exposure (AUC) at steady state.
Approved medical use
Actos is approved as an add-on to diet and exercise for blood sugar control in adults with type 2 diabetes, alone or combined with other diabetes drugs (FDA label). DailyMed also lists generic pioglitazone and fixed-dose combinations with metformin (DailyMed listings).
| Situation | Labeled dosing |
|---|---|
| Starting dose, no heart failure | 15 mg or 30 mg once daily, with or without food |
| Starting dose, NYHA class I or II heart failure | 15 mg once daily |
| Titration | Steps of 15 mg based on HbA1c, up to 45 mg once daily |
| With gemfibrozil or another strong CYP2C8 inhibitor | Maximum 15 mg daily |
| With insulin, if low blood sugar occurs | Reduce the insulin dose by 10–25% |
| Children | Not recommended |
The label asks for liver tests (ALT, AST, alkaline phosphatase, bilirubin) before starting, but not routine repeat testing in people without liver disease.
Pioglitazone is not FDA-approved for MASH/NASH, stroke prevention or prediabetes. The trials below in those settings are research uses.
What the research shows
Blood sugar control
In the label’s 26-week dose-ranging trial (408 patients randomized), HbA1c fell relative to placebo by 1.0 percentage point with 15 mg and 30 mg and by 1.6 points with 45 mg. As an add-on to metformin for 16 weeks (328 patients), 30 mg lowered HbA1c by 0.8 points more than placebo (FDA label).
| Trial (label) and arm | Baseline HbA1c | HbA1c difference vs placebo (95% CI) |
|---|---|---|
| 26-week monotherapy, 15 mg | 10.2% | −1.0 (−1.6 to −0.4) |
| 26-week monotherapy, 30 mg | 10.2% | −1.0 (−1.6 to −0.4) |
| 26-week monotherapy, 45 mg | 10.3% | −1.6 (−2.2 to −1.0) |
| 16-week add-on to metformin, 30 mg | 9.9% | −0.8 (−1.2 to −0.5) |
In the same 26-week trial, triglycerides fell about 9% on pioglitazone versus a 4.8% rise on placebo, and HDL cholesterol rose 12–19% versus 8.1%. LDL and total cholesterol did not change consistently.
PROactive: cardiovascular outcomes in type 2 diabetes
PROactive randomized 5,238 people with type 2 diabetes and existing large-vessel (macrovascular) disease to pioglitazone titrated from 15 to 45 mg or placebo, on top of their usual care, for an average of 34.5 months (Dormandy et al., Lancet 2005).
| Outcome | Pioglitazone (n=2,605) | Placebo (n=2,633) | Hazard ratio (95% CI) |
|---|---|---|---|
| Primary composite (death, MI, stroke, ACS, coronary or leg revascularization, amputation) | 514 | 572 | 0.90 (0.80–1.02), p=0.095 |
| Main secondary: death, nonfatal MI or stroke | 301 | 358 | 0.84 (0.72–0.98), p=0.027 |
| Hospitalized heart failure (label) | 149 (5.7%) | 108 (4.1%) | — |
| Fatal heart failure (label) | 25 (1.0%) | 22 (0.8%) | — |
The primary endpoint was not met. The FDA label states that no clinical study has established conclusive evidence of macrovascular risk reduction with Actos. A 2007 meta-analysis of 19 trials (16,390 patients) found fewer deaths, heart attacks or strokes with pioglitazone (HR 0.82, 95% CI 0.72–0.94) but more serious heart failure (HR 1.41, 95% CI 1.14–1.76), without a rise in heart-failure deaths (Lincoff et al., JAMA 2007).
IRIS: insulin resistance after stroke, without diabetes
IRIS randomized 3,876 people who had a recent ischemic stroke or transient ischemic attack (TIA, a “mini-stroke”) and insulin resistance (HOMA-IR over 3.0) but no diabetes. They received pioglitazone (target 45 mg daily) or placebo and were followed for 4.8 years (Kernan et al., NEJM 2016).
| Outcome | Pioglitazone | Placebo | Effect |
|---|---|---|---|
| Stroke or MI (primary) | 9.0% | 11.8% | HR 0.76 (95% CI 0.62–0.93) |
| New diabetes | 3.8% | 7.7% | HR 0.48 (95% CI 0.33–0.69) |
| All-cause death | — | — | HR 0.93 (95% CI 0.73–1.17), not significant |
| Weight gain over 4.5 kg | 52.2% | 33.7% | p under 0.001 |
| Edema | 35.6% | 24.9% | p under 0.001 |
| Fracture needing surgery or hospital care | 5.1% | 3.2% | p=0.003 |
IRIS is the clearest example of the drug’s trade-off: fewer vascular events and less diabetes, at the cost of weight gain, swelling and fractures.
Fatty liver disease (MASH/NASH)
Several randomized trials have tested pioglitazone in nonalcoholic steatohepatitis, now called metabolic dysfunction-associated steatohepatitis (MASH):
| Trial | Population, n | Dose and duration | Main result |
|---|---|---|---|
| Belfort et al., NEJM 2006 | Prediabetes or type 2 diabetes with biopsy-proven NASH, 55 | 45 mg/day + calorie-reduced diet vs diet + placebo, 6 months | Liver fat −54% vs 0%; ALT −58% vs −34%; better steatosis, ballooning and inflammation; fibrosis not significantly different (p=0.08) |
| PIVENS, Sanyal et al., NEJM 2010 | NASH without diabetes, 247 | 30 mg/day vs vitamin E 800 IU vs placebo, 96 weeks | Primary histology endpoint 34% vs 19% placebo (p=0.04, missed the preset p under 0.025); vitamin E 43% (p=0.001); fibrosis not improved |
| Cusi et al., Ann Intern Med 2016 | Prediabetes or type 2 diabetes with NASH, 101 | 45 mg/day vs placebo, 18 months, then 18 months open label | Primary endpoint 58% on pioglitazone, a 41-point difference (95% CI 23–59); NASH resolution difference 32 points; weight +2.5 kg vs placebo |
A 2017 meta-analysis of 8 trials (516 patients) found that thiazolidinediones improved advanced fibrosis (OR 3.15, 95% CI 1.25–7.93) and NASH resolution (OR 3.22, 95% CI 2.17–4.79), with all effects accounted for by pioglitazone. Body weight rose by an average of 2.7% (Musso et al., JAMA Intern Med 2017). The authors noted it was still unknown whether this translates into fewer clinical liver events.
The 2023 AASLD practice guidance says pioglitazone improves NASH and can be considered for patients with NASH and type 2 diabetes. It also states that available data on pioglitazone do not demonstrate an antifibrotic benefit, and notes that its use has been overtaken by GLP-1 receptor agonists and SGLT2 inhibitors, which bring weight loss (Rinella et al., Hepatology 2023). See SGLT2 inhibitors, dapagliflozin and semaglutide.
Use outside medicine
Pioglitazone comes up in bodybuilding discussions as an insulin sensitizer. No controlled trials have tested it in healthy, lean or athletic people, and none has measured muscle gain or performance. The human data come from people with diabetes, prediabetes, insulin resistance after stroke, or fatty liver disease, and in those groups the drug consistently increased body weight and fat mass.
Side effects and risks
Boxed warning: heart failure
Thiazolidinediones cause dose-related fluid retention, which can lead to or worsen congestive heart failure. The risk is highest when pioglitazone is combined with insulin. The label says (FDA label):
- watch for rapid weight gain, shortness of breath or swelling after starting or increasing the dose;
- it is not recommended in symptomatic heart failure;
- starting it in NYHA class III or IV heart failure is contraindicated.
In a 24-week label trial in people with NYHA class II or early class III heart failure, overnight hospitalization for worsening heart failure occurred in 9.9% on pioglitazone versus 4.7% on glyburide. Cardiovascular deaths were 1.9% versus 2.3%.
Edema and weight gain
Edema is dose-related and is more common with insulin. It reverses when the drug is stopped (FDA label).
| Setting (label) | Placebo | 15 mg | 30 mg | 45 mg |
|---|---|---|---|---|
| Edema, monotherapy 16–26 weeks | 1.2% | 2.5% | 4.7% | 6.5% |
| Edema, add-on to insulin 16–24 weeks | 7.0% | 12.6% | 20.5% | 26.1% |
| Median weight change, monotherapy | −1.4 kg | +0.9 kg | +1.0 kg | +2.6 kg |
| Median weight change, add-on to insulin | +0.2 kg | +2.3 kg | +3.3 kg | +4.1 kg |
In PROactive, edema occurred in 27.3% on pioglitazone versus 15.9% on placebo, and median weight rose 3.6 kg versus falling 0.5 kg. The label says the weight gain probably reflects both fluid retention and fat accumulation. Hemoglobin also dropped by 2–4% in monotherapy trials, which the label links to a larger plasma volume.
Bladder cancer
Male rats given pioglitazone for two years developed bladder tumors. In PROactive, bladder cancer was diagnosed in 14 of 2,605 people on pioglitazone (0.54%) and 5 of 2,633 on placebo (0.19%). Over 13 years of trial plus observational follow-up, rates did not differ (HR 1.00, 95% CI 0.59–1.72) (FDA label).
Observational studies disagree. A large Kaiser Permanente cohort of 193,099 people with diabetes found no significant increase with ever-use (HR 1.06, 95% CI 0.89–1.26), although a risk could not be excluded (Lewis et al., JAMA 2015). The label also cites a UK study that did find an association (HR 1.63, 95% CI 1.22–2.19). The label says pioglitazone should not be used in active bladder cancer, and the unknown recurrence risk should be weighed in people with a past bladder cancer.
Fractures
In PROactive, fractures occurred in 5.1% of women on pioglitazone versus 2.5% on placebo; most were outside the spine, in the lower limb or the lower arm (forearm, wrist or hand). Men showed no increase (1.7% vs 2.1%) (FDA label). A meta-analysis of 10 randomized trials of both TZDs found about double the fracture risk in women (OR 2.23, 95% CI 1.65–3.01) but not in men (OR 1.00) (Loke et al., CMAJ 2009). In IRIS, which included men and women without diabetes, serious fractures were 5.1% versus 3.2%.
Other warnings
From the label (FDA label):
- Low blood sugar when combined with insulin or sulfonylureas. In 24-week add-on to insulin trials, hypoglycemia was reported in 43.5% on 30 mg and 47.8% on 45 mg.
- Liver. There are postmarketing reports of fatal and nonfatal liver failure, although controlled trials showed no drug-induced liver injury. The drug should be stopped if ALT rises over 3 times the upper limit without another cause.
- Macular edema (swelling in the back of the eye), reported after marketing, usually in people who also had leg swelling.
- Ovulation. It may restart ovulation in some premenopausal women who were not ovulating, which can lead to unplanned pregnancy.
- Common side effects (5% or more): upper respiratory infection, headache, sinusitis, muscle pain and sore throat.
Interactions
| Drug | Effect on pioglitazone (label) |
|---|---|
| Gemfibrozil (strong CYP2C8 inhibitor) | AUC ≈3.2-fold higher; half-life 8.3 → 22.7 h; cap dose at 15 mg |
| Rifampin (CYP2C8 inducer) | AUC 54% lower |
| Ketoconazole | AUC 34% higher |
| Topiramate | Lower exposure to pioglitazone and its metabolites |
| Insulin or sulfonylureas | Higher hypoglycemia risk; lower their dose if needed |
Who should not use it
- People with NYHA class III or IV heart failure (starting it is contraindicated) and, per the label, anyone with symptomatic heart failure.
- People with active bladder cancer.
- People with type 1 diabetes or diabetic ketoacidosis.
- Children (not recommended).
- Anyone with a known allergy to pioglitazone.
Legal status and anti-doping
Actos and generic pioglitazone are prescription-only drugs in the US (FDA label). Pioglitazone is not named on the WADA 2026 Prohibited List. The list’s metabolic modulators section (S4.4) names AMPK activators, PPAR-delta agonists such as GW1516, insulins and insulin mimetics, among others, but not PPAR-gamma agonists (WADA 2026 Prohibited List).
Key takeaways
- An insulin sensitizer, not an insulin booster. Pioglitazone activates PPAR-gamma and needs the body’s own insulin to work; it lowers both blood sugar and insulin levels.
- Heart failure is the boxed warning. Fluid retention can cause or worsen heart failure, especially with insulin. Starting it in NYHA class III or IV heart failure is contraindicated.
- Weight gain is expected. Median gains at 45 mg were 1.8–4.1 kg in label trials, from both fluid and fat. A small uncontrolled study found fat shifting from visceral to subcutaneous stores.
- Mixed cardiovascular evidence. PROactive missed its primary endpoint, while IRIS cut stroke or MI by 24% in insulin-resistant stroke patients without diabetes. Both showed more edema, weight gain or heart failure.
- Liver benefits, not a liver approval. Trials show improvement in NASH/MASH histology, and AASLD guidance allows its use in NASH with type 2 diabetes, but it is not FDA-approved for liver disease.
- Bone and bladder signals. Fractures roughly double in women, and bladder cancer data remain inconclusive enough that the label advises against use in active bladder cancer.
Sources
- Takeda Pharmaceuticals America. ACTOS (pioglitazone) tablets prescribing information. Revised March 2025. DailyMed.
- U.S. National Library of Medicine. DailyMed: pioglitazone product labels. DailyMed.
- Dormandy JA, Charbonnel B, Eckland DJ, et al. Secondary prevention of macrovascular events in patients with type 2 diabetes in the PROactive Study (PROspective pioglitAzone Clinical Trial In macroVascular Events): a randomised controlled trial. Lancet. 2005;366(9493):1279–1289. PubMed.
- Lincoff AM, Wolski K, Nicholls SJ, Nissen SE. Pioglitazone and risk of cardiovascular events in patients with type 2 diabetes mellitus: a meta-analysis of randomized trials. JAMA. 2007;298(10):1180–1188. PubMed.
- Kernan WN, Viscoli CM, Furie KL, et al. Pioglitazone after ischemic stroke or transient ischemic attack. N Engl J Med. 2016;374(14):1321–1331. PubMed.
- Miyazaki Y, Mahankali A, Matsuda M, et al. Effect of pioglitazone on abdominal fat distribution and insulin sensitivity in type 2 diabetic patients. J Clin Endocrinol Metab. 2002;87(6):2784–2791. PubMed.
- Belfort R, Harrison SA, Brown K, et al. A placebo-controlled trial of pioglitazone in subjects with nonalcoholic steatohepatitis. N Engl J Med. 2006;355(22):2297–2307. PubMed.
- Sanyal AJ, Chalasani N, Kowdley KV, et al. Pioglitazone, vitamin E, or placebo for nonalcoholic steatohepatitis. N Engl J Med. 2010;362(18):1675–1685. PubMed.
- Cusi K, Orsak B, Bril F, et al. Long-term pioglitazone treatment for patients with nonalcoholic steatohepatitis and prediabetes or type 2 diabetes mellitus: a randomized trial. Ann Intern Med. 2016;165(5):305–315. PubMed.
- Musso G, Cassader M, Paschetta E, Gambino R. Thiazolidinediones and advanced liver fibrosis in nonalcoholic steatohepatitis: a meta-analysis. JAMA Intern Med. 2017;177(5):633–640. PubMed.
- Rinella ME, Neuschwander-Tetri BA, Siddiqui MS, et al. AASLD Practice Guidance on the clinical assessment and management of nonalcoholic fatty liver disease. Hepatology. 2023;77(5):1797–1835. PubMed.
- Lewis JD, Habel LA, Quesenberry CP, et al. Pioglitazone use and risk of bladder cancer and other common cancers in persons with diabetes. JAMA. 2015;314(3):265–277. PubMed.
- Loke YK, Singh S, Furberg CD. Long-term use of thiazolidinediones and fractures in type 2 diabetes: a meta-analysis. CMAJ. 2009;180(1):32–39. PubMed.
- World Anti-Doping Agency. The 2026 Prohibited List: International Standard. In effect January 1, 2026. WADA 2026 Prohibited List.