Statin Induced New-onset Diabetes Mellitus – A Narrative Review

Abstract

Statins are widely used to lower LDL cholesterol and prevent cardiovascular disease. Although generally safe, there is increasing evidence that they may increase the risk of new-onset diabetes mellitus (NODM). Recent studies show a modest but consistent rise in diabetes risk associated with statin use, especially among individuals with prediabetes, obesity, or other metabolic risk factors. High-intensity statins like atorvastatin and rosuvastatin carry a greater risk compared to agents such as pravastatin or pitavastatin. Proposed mechanisms include increased insulin resistance, impaired insulin secretion, and changes in hepatic glucose production. Despite the diabetes risk associated with statins, their well-established cardiovascular benefits generally outweigh this concern. This review discusses the current evidence on the association between statin use and NODM, explores potential biological mechanisms, compares the diabetogenic effects of different statins, assess the impact of statin dose and intensity, and outlines implications for clinical practice.

1. Introduction

Statins play a key role in both primary and secondary prevention of cardiovascular events by lowering low-density lipoprotein cholesterol (LDL-C), a major contributor to atherosclerosis.1 Statins are the most commonly prescribed medications for lowering cholesterol and are well-established in reducing the risk of atherosclerotic cardiovascular disease (ASCVD). These medications help stabilize existing plaques, reduce inflammation and improve endothelial function.2

Statins are typically safe and well-tolerated. The most frequent side effect is myalgia, affecting 1–10% of users, while rhabdomyolysis is rare but serious condition, occurring in less than 0.1% of cases. Mild elevations in liver enzymes occur in up to 1% of patients.3

However, there is increasing evidence from recent studies and clinical trials, that statin use may be associated with a risk of developing NODM. While the absolute risk is moderate and often outweighed by the cardiovascular benefits,4 it becomes more clinically significant in individuals using statins for primary prevention, where new-onset diabetes may lead to long-term health consequences. In such cases, clinicians may individualize treatment by adjusting the statin dose or choosing agents with lower diabetogenic potential.

This review aims to provide a comprehensive analysis of the association between statin use and NODM. It will explore the available clinical and experimental evidence, discusses proposed mechanisms of statin-induced diabetes, compares diabetogenic potential of different statins, the impact of statin dose and intensity on diabetes risk and considers practical implications for clinical decision-making.

2. Statins and their therapeutic role

Statins are classified based on lipophilicity into lipophilic and hydrophilic agents. Lipophilic statins include atorvastatin, simvastatin, lovastatin, fluvastatin, and pitavastatin, while pravastatin and rosuvastatin are hydrophilic. They are also categorized by potency: high-potency statins (rosuvastatin, atorvastatin), moderate-potency statins (simvastatin, pitavastatin), and low-potency statins (pravastatin, fluvastatin, lovastatin).5

The 2018 AHA/ACC Cholesterol Clinical Practice Guidelines recommend moderate- to high-intensity statin therapy based on individual ASCVD risk categories. For adults at high risk (10-year ASCVD risk of >20%), a reduction of 50% or greater in LDL-C is recommended, typically achieved with high-intensity statin therapy. For adults at intermediate risk (10-year ASCVD risk between 7.5% and 20%), moderate-intensity statin therapy is advised to lower LDL-C levels by at least 30%, with 50% reduction for those nearing high risk. Adults aged 40–75 years with diabetes are advised to undergo moderate-intensity statin therapy, regardless of ASCVD risk. For individuals aged 20–75 years with LDL-C levels of 190 mg/dL (4.9 mmol/L) or higher, maximally tolerated statin therapy is recommended. In borderline risk (10-year ASCVD risk between 5% and 7.5%), the presence of risk-enhancing factors may justify the initiation of moderate-intensity statin therapy after a discussion between the clinician and patient.6

The U.S. Preventive Services Task Force (USPSTF) recommends statins for primary prevention of cardiovascular disease (CVD) in adults aged 40–75 years who have at least one CVD risk factor (e.g., dyslipidemia, diabetes, hypertension, or smoking) and an estimated 10-year CVD risk ≥10%. A review of 22 trials, showed statins significantly reduced all-cause mortality (n = 85,816; relative risk [RR], 0.92; absolute risk difference [ARD], −0.35%), fatal or nonfatal stroke (n = 76,610; RR, 0.78; ARD, −0.39%), fatal or nonfatal myocardial infarction (n = 76,498; RR, 0.67; ARD, −0.89%).1

3. Epidemiological evidence and risk assessment of statin-induced diabetes mellitus

7The potential link between statins and NODM was first observed in 1995 in the West of Scotland Coronary Prevention Study (WOSCOPS), where 139 out of 5974 participants without diabetes at baseline developed type 2 diabetes during follow-up. 7 Since then, numerous studies have consistently shown that statin use is linked to an increased risk of developing NODM, especially in people receiving high-intensity statins or those with existing risk factors.8 The risk of statin-induced diabetes appears to be influenced by several factors such as the type of statin used, dose and intensity of therapy, degree of LDL-C reduction, and patient's baseline metabolic status.

Various meta-analysis report a 9–12% relative increase in diabetes risk,7–9 while observational studies suggest even higher rates (55% increased risk).10 The Women's Health Initiative found a 48% increased risk in postmenopausal women taking statins.11,12 Clinical trials such as JUPITER and SPARCL showed a 25–34% higher risk of diabetes with rosuvastatin and atorvastatin, respectively.9,13 In a Japanese cohort study, the 5- and 10-year diabetes incidence among statin users was 13.6% and 15.6%, with hazard ratios of 1.66 and 1.61, supported by Kaplan-Meier curves (P < 0.001).14 Long-term data show higher diabetes rates in statin users (9.3 per 1000 person-years) compared to non-users (6.13 per 1000 person-years).15

Although the increase in diabetes risk is moderate, estimates suggest an absolute risk increase of 0.1%–0.5%, equating to one additional case per 100 to 255 statin users over five years16,17 - it is clinically relevant. For every 10 major cardiovascular events prevented by statins, there may be one extra case of diabetes.17 Another estimate noted one additional case per 137 users when LDL-C was reduced by 30–40%, and one per 108 users when reduction was 40–50%.18 These findings indicate a higher risk with more intensive statin therapy and highlights the need for individualized treatment decisions based on each patient's overall risk profile.

4. Proposed mechanism of statin induced diabetogenecity

Multiple studies have proposed that statin associated dysglycemia results from combination of increased insulin resistance, impaired pancreatic β-cell function and insulin secretion, enhanced hepatic gluconeogenesis and changes in glucose uptake and transport.16 Statin therapy has been shown to reduce insulin sensitivity by 24% and reduction in insulin release by 12%, increasing the risk of NODM.19

4.1. Predisposition to insulin resistance

In patients using statin, reduced cellular responsiveness to insulin occurs through hydroxy-3-methylglutaryl-coenzyme A (HMG CoA) reductase enzyme inhibition.20 By inhibiting HMG-CoA reductase, statins reduce the synthesis of isoprenoids such as farnesyl pyrophosphate and geranylgeranyl pyrophosphate (GGPP), which are critical for protein prenylation and insulin signalling.7,8

Depletion of these isoprenoids impairs trafficking of glucose transporter type 4 (GLUT4) transporters and weakens activation of the phosphoinositide 3-kinase – protein kinase B (PI3K-AKT) pathway, thereby reducing glucose uptake. In addition, low isoprenoid levels activate adipose tissue macrophages and the NOD like receptor protein 3 (NLRP3) inflammasome, leading to increased production of interleukin-1β (IL-1β).21,22

Increased IL-1β disrupts insulin signalling by promoting inhibitory serine phosphorylation and degradation of insulin receptor substrate-1(IRS-1), while also stimulating nuclear factor kappa-lightchain-enhancer of activated B cells (NF-κB) mediated inflammatory cascades that enhance IL-1β release, creating a viscous cycle. Prolonged exposure to IL-1β further decreases IRS-1 expression, reinforcing a cycle of inflammation and impaired insulin signalling that contributes to insulin resistance22,23 (Fig. 1).

Fig. 1.

Flow diagram illustrating the proposed mechanism of statin-induced insulin resistance. Abbreviations: IRS-1 - Insulin Receptor Substrate-1; HMG-CoA – 3-Hydroxy-3-Methylglutaryl-Coenzyme A; JNK – c-Jun N-terminal Kinase; IL-1β – Interleukin-1 Beta; NF-κB – Nuclear Factor Kappa-Light-Chain-Enhancer of Activated B Cells; GLUT4 – Glucose Transporter Type 4.

4.2. Impairing pancreatic beta cell function and insulin secretion

Statins can impair pancreatic beta cell function and insulin secretion through multiple pathways-

1. Statins upregulate LDL receptors in tissues, including pancreatic beta cells, causing increased LDL-C accumulation inside these cells. This accumulation impairs glucose transporter type 2 (GLUT2) function and voltage-gated calcium channels, leading to decreased insulin secretion. Increased LDL-C in beta cells also causes mitochondrial dysfunction and oxidative stress leading to cell death.24

2. Decreased isoprenoids by statin reduces coenzyme Q10 (CoQ10) production, which is essential for the electron transport chain. Lower CoQ10 level leads to decreased ATP production, keeping the potassium channel open and disrupts insulin secretion. Additionally, low CoQ10 causes oxidative stress and beta cell death.24

3. Chronically elevated IL-1β damages pancreatic beta cells by inducing oxidative stress through reactive oxygen species production, activating nitric oxide pathway leading to mitochondrial dysfunction,25 impairing voltage gated calcium channels affecting insulin secretion, and directly activating apoptosis via mitochondrial damage leading to reduced beta cell mass25 (Fig. 2).

Fig. 2.

Flow diagram illustrating the proposed mechanism of statin-induced impairment of insulin secretion. Abbreviations: LDL – Low-Density Lipoprotein; IL-1β – Interleukin-1 Beta; CoQ10 – Coenzyme Q10; ATP – Adenosine Triphosphate; GLUT2 – Glucose Transporter Type 2; ROS – Reactive oxygen species.

4.3. Hepatic gluconeogenesis and glucagon dysregulation

Statins decrease the production of isoprenoids such as GGPP, which leads to upregulation of microRNA-495 and subsequent downregulation of sirtuin 6 (Sirt6). Reduced Sirt6 expression increases the activity of forkhead box protein O1 (FoxO1), enhancing the transcription of phosphoenolpyruvate carboxykinase (PEPCK) and glucose-6-phosphatase. This upregulation promotes gluconeogenesis in hepatic cells, ultimately resulting in fasting hyperglycaemia26 (Fig. 3).

Fig. 3.

Flow diagram illustrating the proposed mechanism of statin-induced increase in hepatic gluconeogenesis. Abbreviations: Sirt6 – Sirtuin 6; FoxO1 – Forkhead Box Protein O1; PEPCK – Phosphoenolpyruvate Carboxykinase.

4.4. Other proposed mechanism

Statins can reduce adiponectin, leptin, insulinlike growth factors and GLP-1 secretion leading to decreased insulin sensitivity and impaired glucose uptake.16,22 They also alter gut microbiota composition, affecting bile acid metabolism leading to reduced FXR activation, which contributes to increased hepatic gluconeogenesis and insulin resistance.4 Elevated mitochondrial short-chain acylcarnitine associated with statin use can impair IRS-1 phosphorylation and activate stress kinases like c-jun N-terminal kinase (JNK) and NF-κB7. Statins have also been shown to reduce Sirt6 expression and increase FoxO1 activity, both of which may contribute to higher blood sugar levels.26 Atorvastatin has been shown to impair insulin's ability to suppress glucagon secretion from alpha cells, causing postprandial hyperglycaemia.16 Rosuvastatin may activate the mechanistic Target of Rapamycin (mTOR) pathway, inhibiting autophagy which leads to mitochondrial damage, cellular stress, and reduced insulin secretion.27

5. Impact of statin dose and potency on new onset diabetes risk

Numerous studies and meta-analyses indicate dose-dependent link between statin usage and NODM. Higher doses and more intense statin treatments are consistently linked to greater risk compared to moderate or low doses. High-intensity statins are associated with 36% higher proportional risk of diabetes (RR 1.36), equating to an absolute excess of 1.27% annually, while low- to moderate-intensity statins show only 10% increase (RR 1.10).9

Specific statins like atorvastatin, simvastatin, and rosuvastatin exhibit dose-dependent impacts on glucose metabolism.19,28,29 Clinical research has demonstrated that higher doses of atorvastatin (e.g., 80 mg) correlate with elevated HbA1c levels and 34% increased risk of diabetes.29 In vitro studies on INS-1E rat insulinoma cells revealed that rosuvastatin dose-dependently reduced insulin secretion and β-cell viability.27 Simvastatin has been found to downregulate Sirt6 expression in hepatocytes in a concentration-dependent manner, affecting insulin sensitivity.26

Meta-analytical data further support the dose-response relationship. A meta-analysis by Preiss et al. identified 12% increased risk of NODM with intensive-dose statin therapy compared to moderate-dose therapy (pooled OR 1.12; 95% CI 1.01–1.24).30 Carter et al. observed increased risk at both moderate (HR 1.22) and high doses (HR 1.30) compared to low doses.24 In terms of absolute risk, treating 498 patients with high-intensity statins for one year, compared to moderate doses, results in one additional case of diabetes.30

One study have found that greater reductions in LDL cholesterol are linked to higher diabetes risk, with 33% increase when LDL-C was ≤1.8 mmol/L and 16% increase when LDL-C was between 1.8 and 2.59 mmol/L.31 Genetic evidence also supports these findings. Variants in the HMG-CoA reductase (HMGCR) gene that result in greater LDL-C reduction were linked to higher risk of type 2 diabetes, 32 suggesting that the diabetogenic effect of statins may be related to the extent of LDL-C lowering and statin potency.

The duration of statin exposure is another crucial factor. In a study, individuals using statins for more than two years had hazard ratio of 3.33 (95% CI 1.84–6.01) for developing diabetes, whereas shorter durations did not show statistically significant risk.33

6. Heterogeneity in diabetes risk among individual statins

The type of statin can significantly influence the risk of developing NODM. Statins, such as pravastatin and pitavastatin, are typically linked to reduced risk of NODM.8,24 Pravastatin has been found to have minimal effects on glucose metabolism and may even enhance insulin sensitivity in individuals with normal insulin function.19 Pitavastatin has shown neutral or positive effects on insulin sensitivity, improving HOMA-IR in healthy individuals and maintaining stability in those with insulin resistance.34 Conversely, atorvastatin and simvastatin are more often associated with negative impacts on glucose metabolism, including impaired insulin secretion and increased insulin resistance in both healthy and insulin-resistant individuals.19,22 Rosuvastatin's strong LDL-lowering ability has been linked to increased HbA1c and HOMA-IR levels, likely explaining its relatively higher potential to induce diabetes.34

6.1. Comparative risk across statins

Evidence from meta-analyses and large observational studies indicates that the likelihood of developing NODM varies among different statins. High-intensity atorvastatin (80 mg) has consistently been linked to the greatest risk, with odds ratios (OR) ranging from 1.15 to 1.34. Rosuvastatin has also shown a significant association, with ORs between 1.18 and 1.25.28,29 In a clinical trial, both atorvastatin 80 mg/day and rosuvastatin 40 mg/day led to an average increase in HbA1c of about 0.3% over 18 weeks.12 Some analyses identified rosuvastatin as the highest-risk agent, with crude incidence rates of 34.2 per 1000 person-years, followed closely by atorvastatin with 30.7 per 1000 person-years. 24 At moderate doses, rosuvastatin 10 mg/day carried the highest risk (OR 1.11), followed by atorvastatin 10 mg/day (OR 1.04), while pravastatin 20 mg/day had almost neutral effect (OR 0.99).28 Direct comparisons consistently showed pravastatin as the lowest-risk agent. The estimated number needed to harm (NNTH) was 172 for atorvastatin, 210 for rosuvastatin, and 363 for simvastatin, all compared to pravastatin.24

Simvastatin carried an intermediate risk, with hazard ratios (HR) ranging from 1.10 to 1.14. In contrast, lovastatin and fluvastatin generally did not show a significant association with diabetes in most studies.8,24,28 Crude incidence rates per 1000 person-years were 26.22 for simvastatin, 22.64 for pravastatin, 21.80 for lovastatin, and 21.52 for fluvastatin.24 Pitavastatin consistently showed the most favourable profile. In multiple cohort studies, its incidence of NODM was substantially lower (3.0–12.7%) compared with atorvastatin (8.4–18.3%) and rosuvastatin (10.4–21.6%), with hazard ratios around 0.7 relative to both atorvastatin and rosuvastatin.20,35,36

In summary, atorvastatin and rosuvastatin are associated with the highest risk of NODM, while pitavastatin and pravastatin are associated with the lowest risk, and simvastatin falls in between. Fluvastatin and lovastatin are generally considered neutral in terms of diabetes risk (Table 1).

Table 1.

Diabetogenicity of individual statins.

Statin	Relative Risk(OR/HR)	Crude Incidence per 1000 person-years	Risk Category
Rosuvastatin	OR 1.18–1.25	34.21	High risk across studies (NNTH - 210 relative to pravastatin
Atorvastatin	OR 1.15–1.34	30.70	High risk across studies (NNTH - 172 relative to pravastatin)
Simvastatin	HR 1.10–1.14	26.22	Moderate risk (NNTH - 363 relative to pravastatin)
Pitavastatin	HR - 0.69 vs atorvastatin HR - 0.74 vs rosuvastatin HR - 0.72 vs combined atorvastatin and rosuvastatin		Consistently low risk across studies
Pravastatin	OR 0.99	22.64	Low risk across studies; Some isolated studies showed higher risk, but these were limited by small sample sizes.
Fluvastatin	No significant risk	21.5	Neutral
Lovastatin	No significant risk	21.8	Neutral

Comparative risk estimates assessing new-onset diabetes among statin users. Crude incidence rates are expressed per 1000 person- years. Abbreviations: OR – Odds Ratio; HR – Hazard Ratio; NNTH – Number Needed to Harm.

7. Comorbid conditions influencing statins diabetogenecity

The likelihood of developing new-onset diabetes mellitus due to statin use is notably higher in people with existing metabolic risk factors. Studies has consistently indicated that transition to diabetes is more prevalent among statin users who are obese, older, female, Asian, or have hypertension, hypertriglyceridemia, impaired fasting glucose, or metabolic syndrome. Other factors that elevate the risk include high baseline HbA1c, family history of diabetes and chronic kidney disease.4,8,12,20,22

Statin users with borderline glycemic markers, such as fasting blood glucose over 100 mg/dL or HbA1c above 6% - are especially at risk.8,33 The risk is further heightened in individuals with body mass index (BMI) of 30 kg/m² or higher, triglycerides of 150 mg/dL or more, or multiple metabolic abnormalities.8,16

7.1. 4-Tier model for risk stratification of NODM risk based on comorbidities by Kohli P et al

Lowest risk

Normal fasting glucose and either triglycerides ≤150 mg/dL or BMI <27 kg/m².

Intermediate risk

Elevated triglycerides (normal - 150 mg/dL) or BMI (normal - 18.5–24.9 kg/m²) alone.

Higher risk

Prediabetes alone.

Highest risk

Prediabetes with elevated triglycerides (normal - 150 mg/dL) or BMI (normal - 18.5–24.9 kg/m²), with an incidence more than five times higher than the lowest risk group.37

In one cohort, prediabetic statin users showed 28.5% incidence of diabetes over 4.1 years, with 20% higher relative risk compared to non-users (HR 1.20).38 In another study, the incidence of diabetes reached 22.8% for those with both prediabetes and triglycerides >150 mg/dL, and 20.5% with a BMI ≥27 kg/m². The hazard ratio was 7.0 for both factors combined, 6.7 for prediabetes with high triglycerides, 3.5 for prediabetes alone, and 1.4 for high BMI alone. Conversely, patients with normal fasting glucose and triglycerides ≤150 mg/dL or BMI <27 kg/m² had an incidence of only 2.6%– 2.8%.37

8. Current gap in existing literature

While the association between statin use and NODM is well established, the possibility of reversing this risk has not been thoroughly investigated. Some studies indicate that discontinuing statin might lower the risk of diabetes,12,15,19 but current evidence is not sufficient to support risk reversibility in standard clinical practice. All significant studies agree that cardiovascular benefits of statins outweigh the risk of developing diabetes, and discontinuing statins solely due to this risk is not recommended. However, for patients whose ASCVD risk score significantly decreases after prolonged statin use and lifestyle changes, it is reasonable to consider whether discontinuing statin use could reverse or lower the risk of diabetes. This remains an important area for future research. Alternative therapies to statins, such as ezetimibe, have shown a neutral effect on glucose metabolism, with no increased risk of NODM reported in trials like IMPROVE-IT.39 More studies are needed to determine whether these agents can provide cardiovascular protection comparable to statins without the associated diabetes risk. Furthermore, given the established connection between statins and diabetes risk, it is clinically important to regularly monitor blood glucose and HbA1c levels. Although most research acknowledges the necessity of such monitoring, there are no definitive guidelines on frequency of this monitoring. Creating evidence-based protocols for glucose monitoring in statin users is another area that warrants further investigations.

9. Conclusion

Statins play a central role in prevention of ASCVD, but recent studies show a significant association between statin use and NODM, particularly with high intensity statins. Although the absolute risk of diabetes is relatively moderate, it becomes more significant in primary prevention scenarios and in patients with borderline blood sugar levels. A personalized approach is needed, weighing the cardiovascular benefits of LDL reduction against diabetes risk. In lower-risk patients, clinicians may consider starting therapy with low-dose or less diabetogenic statins, such as pravastatin or pitavastatin, along with regular monitoring of fasting blood glucose or HbA1c. Educating patients on diet, exercise, and weight control remains essential to minimize dysglycemia. For most patients, the cardiovascular benefits clearly outweigh the diabetes risk, but careful selection of statin type and dose, guided by individual risk factors, will optimize patient outcomes. Future research should focus on refining risk stratification, exploring pharmacogenomics, and conducting long-term studies to balance cardiovascular protection with metabolic safety.