Safety Profile of Statins: An Umbrella Review of Systematic Reviews and Meta-analyses

Abstract

Background:

Statins are a cornerstone of therapy to reduce LDL cholesterol and cardiovascular diseases. However, their use may cause adverse events. This umbrella review aims to comprehensively review the safety profile of statin therapy.

Methods:

PubMed and Embase were searched to identify systematic reviews, scoping reviews, meta-analyses and network meta-analyses on statin safety. Extracted data included study type, data source, number of included studies, search date range, participant numbers, interventions, safety outcomes, results and funding. Risk of bias was assessed using the AMSTAR 2 instrument. Data underwent descriptive statistical analysis.

Results:

Seventy-two reviews were included, mostly based on clinical trials (70.8%) and observational studies (59.7%). Only 21 (29.2%) reported results for individual statins, with atorvastatin being the most frequently evaluated individual statin (95.2%). Nearly 86% of the reviews were rated as Critically Low in overall confidence. Statins were associated with an increased risk of new-onset diabetes, particularly rosuvastatin and atorvastatin, whereas pitavastatin may be protective. Liver injury/dysfunction and elevated transaminases were observed, along with a small risk of muscle-related symptoms, specifically at higher statin intensity. Rare serious events, such as necrotising myopathy, were reported. Pregnancy studies indicated an increased risk of spontaneous abortion.

Conclusion:

Although overall statin safety is favourable, the risks of diabetes, liver dysfunction, enzyme elevations and muscle symptoms require consideration, especially with high-intensity therapy or specific statins. Pitavastatin may be a safer option for patients at metabolic risk. Therapy should be tailored to individual patient factors. Further research is needed to clarify rare or inconsistent outcomes.

Statins, or 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors, are the primary pharmacological choice for treating hypercholesterolaemia and for the primary and secondary prevention of cardiovascular events.1,2 Statin inhibition of HMG-CoA reductase reduces intracellular cholesterol synthesis, upregulating hepatic LDL receptors and increasing LDL cholesterol clearance from the circulation.3 Statins also exhibit pleiotropic effects, including anti-inflammatory properties, plaque stabilisation and improvements in endothelial function, all of which contribute to their overall cardiovascular protective benefits.4

Statins are generally considered well tolerated and safe, but some adverse events (AEs), particularly those associated with long-term use, have been reported.5 The most frequent AEs are statin-associated muscle symptoms. These events can significantly impact patients’ quality of life and often lead to therapy discontinuation or non-adherence.5,6 Other reported AEs include elevation of hepatic enzymes (to more than three times the upper limit of normal) and, more rarely, serious liver injury or hepatotoxicity.5,7

Concerns about new-onset diabetes (NOD) and changes in glycaemia and HbA_1c have also been associated with statin therapy. The risk is heightened in patients with pre-existing risk factors for diabetes and with more intensive statin dosage.5

Neurological events, changes in sex hormone levels, eye disorders, renal events, tendinitis and cancer have also been associated with the use of statins.5,8 However, the available data on these events often present limitations, such as different types of studies, inconsistency in the results, a paucity of studies or difficulties in controlling for confounding variables that may influence the development of AEs independently of statin therapy.5

Despite their benefits, the safety profile of statins remains a subject of ongoing discussion, particularly concerning AEs that lead to therapy discontinuation or impact quality of life and patient adherence. Therefore, a comprehensive review of available evidence assessing the risk of AEs associated with statin therapy is of utmost importance.

The aim of this umbrella review was to comprehensively review the safety profile of statin therapy by reviewing the most recent systematic reviews (SRs) and meta-analyses (MAs).

Methods

The umbrella review was conducted in accordance with the umbrella reviews chapter in the JBI Manual for Evidence Synthesis and is reported in accordance with the PRIOR guidelines for overviews of reviews of healthcare interventions.9,10 It was also reported according to PRISMA 2020 (Supplementary Table 1).11 The study protocol was registered in PROSPERO (CRD420251185582).12

Information Sources and Search Strategy

A systematic search identified relevant publications in PubMed and Embase. A time restriction of 10 years (2016–2025) was applied. The last search was performed in June 2025. Indexed and free-text terms for ‘statins’ and the seven authorised individual statins were used. Filters were applied to specifically identify SRs, scoping reviews, MAs and network MAs. Cerivastatin was not included because it was withdrawn from the market in 2001. Given the requirement for full text, additional searches in the grey literature were not performed. The full search strategy is presented in Supplementary Tables 2 and 3. EndNote 21 (Clarivate Analytics) was used to manage duplicates.13

Eligibility Criteria and Study Selection

The following eligibility criteria were applied:

• Population: statin users

• Intervention: statins, excluding cerivastatin

• Comparator: non-statin, placebo, active treatment, no treatment, non-pharmacological interventions

• Outcomes: any safety outcome

• Types of study: SRs, scoping reviews, MAs and network MAs

• Languages: English and/or Portuguese.

Narrative reviews, conference abstracts, notes and editorials were excluded. The reference lists of the included studies were reviewed to identify additional relevant articles. Results were screened independently by two investigators (AP and DM). Any disagreements were resolved by a third investigator (CA).

Data Items and Collection

Data were independently extracted by two investigators (AP and DM). A Microsoft Excel form was used to extract the following characteristics from each study: reference, type of study, data source, objective, number ofincluded studies, date range ofincluded studies, population/therapeutic indication, number of participants, interventions, comparators, safety outcome, results and funding (Supplementary Table 4).

Safety outcomes were categorised in accordance with ICD-11 for Mortality and Morbidity Statistics at level 1 and according to the verbatim report by the authors of each study.14

Risk of Bias Assessment

The methodological quality of the included reviews was assessed using the AMSTAR 2 (A MeaSurement Tool to Assess systematic Reviews 2) instrument.15 AMSTAR 2 provides an overall rating of confidence in the results of each included review, based on the presence of weakness in critical and non-critical domains.15 The results of the AMSTAR 2 assessment were presented graphically using robvis.16

Data Synthesis

The results were analysed using descriptive statistics.

Results

In all, 6,204 articles were identified through the literature search. After removing duplicates, 4,846 articles were screened against the eligibility criteria. One additional article was identified through hand-searching. A final sample of 72 SRs and/or MAs (reviews) met the eligibility criteria. The study selection process is shown in Figure 1. The included and excluded studies are listed in Supplementary Material 1 and Supplementary Material 2, respectively.

Figure 1: Flow Chart Showing the Identification, Screening and Selection of Systematic Reviews.

Study Characteristics and Quality

Table 1 summarises the characteristics of the included reviews. The most frequently reported type of study was an SR with MA (n=44; 61.1%). Fifty-three (73.6%) studies conducted a literature search from database inception. The number of studies included in each review ranged from three to 2,895. Most of the reviews included clinical trials (n=51; 70.8%) and observational studies (n=43; 59.7%), with cohort (n=35) and case-control studies (n=20) being the most common designs. The number of participants ranged from 80 to 46,728,889. Only 21 (29.2%) studies reported the results individually by statin. The most frequently evaluated statin was atorvastatin (n=20), followed by simvastatin (n=17). The overall confidence in the results of most reviews was rated as Critically Low (n=62; 86.1%). Four SRs demonstrated the highest compliance with methodological quality criteria, with a moderate confidence rating.17–20 None of the four studies complied with the non-critical domain 10.17–20 The study by Teoh et al. also did not comply with the non-critical domain 6.20 Detailed findings for each study are presented in Supplementary Material 3, Supplementary Figure 1 and Supplementary Table 5.

Table 1: Characteristics of the Included Systematic Reviews and Meta-analyses.

Type of Study
Systematic review/meta-analysis	44 (61.1%)
Systematic review without meta-analysis	16 (22.2%)
Systematic review/network meta-analysis	7 (9.7%)
Individual participant data meta-analysis	2 (2.8%)
Systematic review/pairwise meta-analysis/network meta-analysis/dose-response meta-analysis	2 (2.8%)
Scoping review	1 (1.4%)
Date Range of Included Studies
Search conducted from inception	53 (73.6%)
Searches conducted in a defined date range	17 (23.6%)
Minimum range	1 year
Maximum range	68 years
No search conducted	2 (2.8%)
No. Studies
Minimum	3
Maximum	2,895
Data Source
Clinical trial	51 (70.8%)
Clinical trial not otherwise specified	7 (9.7%)
Randomised controlled trial	44 (61.1%)
Observational study	43 (59.7%)
Case report(s)	5 (6.9%)
Case-control study	20 (27.8%)
Cohort study	35 (48.6%)
Cross-sectional study	10 (13.9%)
Nested case-control study	2 (2.8%)
Observational not otherwise specified	4 (5.6%)
Review	6 (8.3%)
Systematic review	2 (2.8%)
Systematic review/meta-analysis	4 (5.6%)
Mini-review	1 (1.4%)
In vivo/in vitro study	1 (1.4%)
No. Participants
Minimum	80
Maximum	46,728,889
Intervention
Statins (as a drug class)	67 (92.9%)
Atorvastatin	2 (2.9%)
Statins according to dose intensity	3 (4.3%)
Presentation of Results by Intervention
Presented by statins as a drug class	51 (70.8%)
Presented by statin individually	21 (29.2%)
Atorvastatin	20 (27.8%)
Fluvastatin	14 (19.4%)
Lovastatin	15 (20.8%)
Pitavastatin	8 (11.1%)
Pravastatin	15 (20.8%)
Rosuvastatin	15 (20.8%)
Simvastatin	17 (23.6%)
Presented by statins in combination with ezetimibe	0 (0%)

Unless indicated otherwise, data are given as n (%). Throughout the table, the percentages were calculated based on the total number of systematic reviews and meta-analyses. In some cases, a systematic review and/or meta-analysis could have more than one data source or intervention.

Safety Profile

The most assessed safety outcomes were ‘05 Endocrine, nutritional or metabolic diseases’ (n=16; 22.2%), including events such as NOD and worsening of glycaemia control, and ‘08 Diseases of the nervous system’ (n=15, 21.4%), including events such as intracerebral haemorrhage and statin-associated necrotising myopathy, followed by ‘15 Diseases of the musculoskeletal system or connective tissue’ (n=12; 16.7%). The classification of safety outcomes according to ICD-11 is shown in Figure 2 and the main results for each ICD-11 level 1 category are described below. The main characteristics of each study, categorised by safety outcome, are presented in Supplementary Table 6.

Figure 2: Safety Outcomes According to ICD-11 Classification.

01 Certain Infectious or Parasitic Diseases

Wu et al. concluded that statins may increase the risk of influenza infection (RR 1.05; 95% CI [1.03–1.07]).21 Similarly, Fan et al. reported a potential association between statin use and an increased risk of herpes zoster infection (OR 1.18; 95% CI [1.11–1.25]).22 However, significant heterogeneity (I²=91.2%; p<0.000) was observed in the analysis of Fan et al. due to regional differences.22

02 Neoplasms

Statins were not associated with an increased risk of cancer. The meta-analysis from Wang et al. (n=47 randomised controlled trials [RCTs]) reported a non-statistically significant overall OR of 1.01 (95% CI [0.93– 1.10]; I²=0%; p=0.767) for cancer incidence comparing statin users with non-users.18 Bolt et al. (n=27 RCTs) found no increased risk of cancer-related mortality (RR 1.03; 95% CI [0.97–1.10]; I²=0%).23 Four SRs/MAs assessed the risk of developing specific type of cancers, reporting no significant association for breast cancer (RR 0.94; 95% CI [0.86–1.03]), a non-significant risk for non-melanoma skin cancer based on RCTs (RR 1.09; 95% CI [0.85–1.39]) and a modestly increased risk based on observational studies (RR 1.11; 95% CI [1.02–1.22]), a potential protective effect in pancreatic cancer (RR 0.84; 95% CI [0.72–0.95]) and a small but significantly increased risk of skin cancer with lovastatin (OR 1.18; 95% CI [1.00–1.39], p=0.048, I²=66.4%) and simvastatin (OR 1.11; 95% CI [1.05– 1.18]; p<0.001; I²=61.8%), although with high heterogeneity.24–27

05 Endocrine, Nutritional or Metabolic Diseases

NOD was the safety outcome most assessed in this category (n=14). Most of the studies (n=7) reported an increased risk of developing NOD associated with statins in general, or with some types of statins, such as rosuvastatin (RR 1.61; 95% CI [1.30–1.98]; I²=81.62%; OR 1.14; 95% CI [1.00– 1.30]; OR 1.17; 95% CI [1.0–1.3], p=0.010; OR 1.17; 95% CI [1.02–1.35]), atorvastatin (RR 1.49; 95% CI [1.31–1.70]; I²=85.28%; OR 1.29; 95% CI [1.0– 1.6]; p=0.042) and atorvastatin 80 mg (OR 1.34; 95% CI [1.14–1.57]).28–34 However, some studies described considerable heterogeneity. Cai et al. reported a potential benefit associated with pitavastatin (OR 0.76; 95% CI [0.61–0.96]).32 Singh et al. also found a lower risk with pitavastatin compared with atorvastatin (RR 0.86; 95% CI [0.79–0.93]) and rosuvastatin (RR 0.77; 95% CI [0.71–0.84]).35

The Cholesterol Treatment Trialists’ Collaboration concluded that both low/moderate-intensity statins (OR 1.10; 95% CI [1.06–1.14]) and high-intensity statins (OR 1.24; 95% CI [1.06–1.44]) increased the risk of worsening of glycaemia.29

08 Diseases of the Nervous System

Six studies assessed the potential association between intracerebral haemorrhage and statin therapy. Li et al. analysed atorvastatin, fluvastatin, lovastatin, pravastatin, rosuvastatin and simvastatin, and found no association.36,37 Bétrisey et al. concluded that there is an increased risk of developing intracerebral haemorrhage with statin therapy.38 Similar results were observed by Pandit et al. (RR 1.53; 95% CI [1.16–2.01], p=0.002; I²=40%) regarding high-dose statin therapy.39 Teoh et al. assessed post-stroke patients and concluded that statin therapy increased the risk of haemorrhagic stroke (RR 1.42; 95% CI [1.07–1.87]; p=0.01; I²=0%) but decreased the risk of ischaemic stroke (RR 0.85; 95% CI [0.75–0.95]; p=0.006; I²=0%).20 Liu et al. and Ziff et al. did not find an association between statin therapy and intracerebral haemorrhage.40,41 Two SRs reviewed case reports on statin-associated necrotising myopathy, and concluded that this event is characterised by proximal muscle weakness, marked creatine kinase elevation, anti-HMG-CoA reductase positivity, and favourable outcomes in most patients following statin discontinuation and immunosuppressive therapy.42,43

09 Diseases of the Visual System

Alves et al. identified an increased risk of cataracts associated with statins (OR 1.11; 95% CI [1.02–1.21]; p=0.017; I²=97.5%).44 However, a low magnitude of effect and high heterogeneity was also found, biasing the certainty of these findings.44 Yu et al. identified an increased risk of cataracts in those using statins when assessing cohort studies (RR 1.13; 95% CI [1.01–1.25]; I²=90.5%) but not when assessing case-control studies and RCTs, concluding that there is no clear evidence to support this association.45

The study of Pan et al. found an increased risk of glaucoma associated with pravastatin (RR 1.20; 95% CI [1.01–1.43]), rosuvastatin (RR 1.23; 95% CI [1.03–1.46]) and simvastatin (RR 1.21; 95% CI [1.02–1.43]).46 Fakhri et al. did not find any statistically significant association between statin use and glaucoma.47

Cai et al. and Wang et al. analysed the association between statins and eye AEs in primary and secondary prevention, respectively.18,32 In the analysis of statins in primary prevention, Cai et al. found that statins were associated with an increased risk of developing eye conditions (OR 1.23; 95% CI [1.04–1.47], I²=0%).32 In the case of statins for secondary prevention, Wang et al. concluded that because of a lack of studies, risk could not be estimated.18

13 Diseases of the Digestive System

Wang et al. also found an increased risk of developing gastrointestinal discomfort with statins for secondary prevention (OR 1.23; 95% CI [1.02–1.48]; I²=25%).18 Cai et al. did not find a significant association of liver injury/dysfunction with pitavastatin (OR 1.04; 95% CI [0.51–2.12]), although significant associations were found for atorvastatin (OR 1.41; 95% CI [1.08–1.85]) and lovastatin (OR 1.81; 95% CI [1.23–2.66]).32 In contrast, Liang et al. reported a significant association of liver injury/dysfunction with fluvastatin (OR 3.50; 95% CI [1.07–11.53]), whereas no significant associations were found for atorvastatin (OR 1.36; 95% CI [0.55–3.39]) or lovastatin (OR 1.89; 95% CI [0.82–4.32]).48 Despite these inconsistencies across individual statins, an association between statin therapy and liver injury/dysfunction was found by both Cai et al. (OR 1.33; 95% CI [1.12–1.58]; I²=0%) and Liang et al. (OR 1.18; 95% CI [1.01–1.39]; I²=0%).32,48

15 Diseases of the Musculoskeletal System or Connective Tissue

Tsui et al. estimated the incidence of muscular adverse reactions by individual statin in the Chinese population, finding that incidence varied between 0.710% (rosuvastatin) and 2.532% (pitavastatin).49 In contrast, Cai et al. found a significant association for rosuvastatin (OR 1.09; 95% CI [1.01–1.16]), but not for pitavastatin (OR 0.70; 95% CI [0.44–1.12]), with muscular adverse reactions.32 Cai et al. and Wang et al. also analysed the association between statins and muscular AEs in primary and secondary prevention, respectively.18,32 In the case of statins for primary prevention, an increased risk was found in self-reported muscular symptoms (OR 1.06; 95% CI [1.01–1.13]; I²=1%) but not for muscular disorders (OR 0.88; 95% CI [0.62–1.24]; I²=0%).32 In the case of statins for secondary prevention, the studies did not show an increased risk in self-reported muscular symptoms (OR 1.00; 95% CI [0.83–1.20]; I²=0%).18

Zhang et al. estimated the risk of osteoarthritis associated with statin therapy and concluded that statins were associated with an increased risk of developing this safety outcome (OR 1.099; 95% CI [1.002–1.206]; p=0.045; I²=79.60%).17 Nonetheless, that SR and MA found a low magnitude of effect and high heterogeneity among studies included.18

Hou et al. identified a small but significant increase in the risk to muscle symptoms associated with statins as a class compared with control treatment (RR 1.05; 95% CI [1.01–1.09]; p=0.406; I²=4.1%).50 When assessed at the level of individual statins, no significant associations were observed for any outcome. In contrast, at a dose level, moderate-intensity statins presented an increased risk of muscle symptoms compared with control treatment (RR 1.13; 95% CI [1.01–1.27]).50 No significant associations were found between statin use and myalgia, myopathy or rhabdomyolysis.50 An individual patient data MA analysed 23 RCTs and assessed multiple muscle events.51 An association with statins, compared with placebo, was found for muscle cramp or spasm (rate ratio 1.09; 95% CI [1.00–1.19]), any muscle pain (rate ratio 1.03; 95% CI [1.01–1.06]) and any muscle pain or weakness (rate ratio 1.03; 95% CI [1.01–1.06]).51 An association was also found for more intensive statin therapy, compared with less intensive statin therapy, for any muscle pain (rate ratio 1.05; 95% CI [1.01–1.09]) and any muscle pain or weakness (rate ratio 1.05; 95% CI [1.01–1.09]).51 Davis and Weller assessed various musculoskeletal events, finding that high-intensity statin therapy presented a higher risk than placebo or moderate-intensity statin therapy, showing a markedly increased risk of elevated creatine kinase.52 No significant risk of rhabdomyolysis was observed in these comparisons.52 No statistically significant differences were found between moderate-intensity statins and placebo for any of the outcomes assessed.52

16 Diseases of the Genitourinary System

Cai et al. and Wang et al. also assessed the association between statins and renal insufficiency in primary and secondary prevention, respectively.18,32 In the case of statins used for primary prevention, Cai et al. found an increased risk for statins as a class (OR 1.14; 95% CI [1.01– 1.28]; I²=0%), and particularly for rosuvastatin (OR 1.13; 95% CI [1.00– 1.28]).32 In the case of statins used for secondary prevention, the MA of Wang et al. did not show an increased risk for the statins evaluated either individually or as a class (OR 0.74; 95% CI [0.18–3.08]; I²=55%).18

18 Pregnancy, Childbirth or the Puerperium

Three SRs/MAs assessed several outcomes, such as spontaneous abortion, stillbirth, elective or induced termination, preterm birth and live birth.19,53,54 All three SRs/MAs estimated an increased risk for spontaneous abortion with statins compared with non-statin strategies.19,53,54 No statistical associations were estimated for the other outcomes, with the exception of live births in the study of Karadas et al. (OR 0.60; 95% CI [0.49–0.75], I²=7%).53

20 Developmental Anomalies

The risk of cardiac malformations associated with the mother’s statin therapy was assessed in three SRs/MAs.19,53,55 Hirsh et al. estimated an OR of 1.4 (95% CI [1.1–1.8]; I²=0%) for statin therapy.19 That review included RCTs and cohort studies and found moderate overall confidence in their results.19 Karadas et al. estimated an adjusted OR of 1.24 (95% CI [0.93– 1.66]; I²=0%) for cardiac malformations.53 That review included cohort and nested case-control studies and reported critically low confidence in their results.53 Vahedian-Azimi et al. estimated an OR of 2.53 (95% CI [0.81, 7.93]; p=0.112) for cardiac malformations in a review that included clinical trials, cohort studies and case reports, and reported critically low confidence in their results.55

21 Symptoms, Signs or Clinical Findings, Not Elsewhere Classified

Five SRs/MAs concluded that there are non-significant changes in HbA_1c and fasting plasma glucose associated with statin therapy.56–60 The SRs/ MAs of Wang et al. and Villani et al. assessed and found an association between statin therapy as a class and elevations in transaminases, with ORs of 1.62 (95% CI [1.20–2.18], I²=69%) and 1.45 (95% CI [1.24–1.69]; p<0.001; I²=14%), respectively.18,61

At the individual statin level, significant differences in changes in HbA_1c and fasting plasma glucose were only seen with atorvastatin, with ORs of 4.0 (95% CI [2.2–7.6]) and 2.66 (95% CI [1.74–4.06]; p<0.001; I²=12%).18,61 Villani et al. also found an association for lovastatin (OR 1.53; 95% CI [1.03–2.28]; p=0.04; I²=0%) and rosuvastatin (OR 1.35; 95% CI [1.06–1.70]; p=0.01; I²=0%) with changes in HbA_1c and fasting plasma glucose.61 Other SRs/MAs did not observe any significant association between statins and changes in HbA_1c and fasting plasma glucose (OR 0.90; 95% CI [0.21– 3.99]; I²=64%).62

Discussion

The present study highlights complex and often conflicting evidence regarding statin safety. Recent research on statin safety has been focused on ‘endocrine, nutritional or metabolic diseases’, ‘diseases of the nervous system’, ‘diseases of the musculoskeletal system or connective tissue’ and ‘symptoms, signs or clinical findings, not elsewhere classified’, which included laboratory examinations such as transaminases.

01 Certain Infectious or Parasitic Diseases

Wu et al. reported an increased risk of developing influenza with statins, but their analysis was limited by few observational studies and the difficulty in controlling confounding.21 Data on this event are conflicting due to pleiotropic effects of statins, such as anti-inflammatory and immunomodulatory effects.63 Vahedian-Azimi et al. described a lower prevalence of influenza among patients receiving statin therapy (OR 0.85; 95% CI [0.73–0.99]; p=0.040), as well as reduced mortality (OR 0.68; 95% CI [0.56–0.82]; p<0.001) among statin-treated patients with influenza.64

Fan et al. reported an increased risk of herpes zoster infection in participants with long-term follow-up.22 However, the analysis was associated with significant heterogeneity.22 It is suggested that the immunomodulatory properties associated with statin therapy may reactivate varicella-zoster virus, leading to herpes zoster infection.22,65 Nonetheless, several confounders, such as comorbidities or concomitant medication, should be controlled in future studies.22,65

02 Neoplasms

Overall, the evidence does not indicate an increased risk of cancer incidence or mortality with statin use, yet findings for specific cancer types are inconsistent. In recent years, statins have gained interest as repurposed anticancer agents.66,67 One of the potential pleiotropic effects of statins includes an antitumour action, by inhibiting the mevalonate pathway.67,68 This impairment affects cell signalling in tumourigenesis.66–68 Lipophilic statins exhibited better results due to broader tissue distribution.66 Due to the inconsistent evidence, large-scale clinical studies may be required to elucidate statin therapy effects on cancer.

05 Endocrine, Nutritional or Metabolic Diseases

NOD associated with statins has been extensively studied. Most SRs suggested an increased risk associated with statins, particularly rosuvastatin and atorvastatin.33,34,58 This effect appears to be dose-dependent, because higher-intensity statin therapy was associated with a greater worsening of glycaemia.29 In contrast, pitavastatin demonstrated a potential protective effect by reducing the risk of NOD.32,33 However, evidence for this effect is limited to a small number of studies. Estimates from both clinical trial-based and observational studies generally show a consistent direction of association, although some observational study estimates appear slightly higher than those reported in RCTs. It should be noted that observational studies are more susceptible to residual confounding related to baseline risk factors, which should be considered when interpreting the magnitude of risk. In addition, although intensity-dependent effects have been reported, these findings should be interpreted with caution because treatment intensity is not always directly randomised and may be influenced by factors such as adherence or dose adjustments.

The potential mechanisms for NOD appeared to be related to the inhibition of HMG-CoA reductase, which can affect glucose homeostasis, leading to impaired β-cell function and decreased insulin sensitivity and secretion.69 Other mechanisms include statin-induced reduction of coenzyme Q10 levels and activation of the NLRP3 inflammasome.69 Long-term studies reported pancreatic exhaustion.70 The impact on glycaemic control may diverge among statin types and doses.69 Potential risk factors and confounders could have an impact on the development of NOD, such as genetic variants, BMI, insulin resistance, total triglycerides, blood pressure and smoking, among others.69,70 Evidence suggests that pitavastatin may exert its beneficial effects by increasing plasma concentrations of adiponectin.71 The pharmacokinetic properties of pitavastatin show higher systemic bioavailability results compared with other statins, which potentially enhance extrahepatic effects, specifically influencing adipose tissue and circulating adiponectin concentrations.71,72 This effect leads to improvement in insulin sensitivity, an increase in β-cell function, a decrease in adipose tissue inflammation and an increase in HDL levels, with potential beneficial effects on NOD.71,72

08 Diseases of the Nervous System

Data on statins and intracerebral haemorrhage are conflicting. Whereas some SRs found an increased risk, particularly for high-intensity therapy, others did not find any significant association. Statins may provide neuroprotective effects.73 These effects stem from their inhibition and regulation of a range of brain processes, such as tissue necrosis, oedema, apoptosis and diminished cerebral blood flow, among others. However, some studies showed that statin therapy may increase the risk of recurrent intracerebral haemorrhage.73 Nonetheless, some studies also found a protective effect in patients with previous ischaemic stroke.37,74,75

Statin-associated necrotising myopathy is a rare and serious AE, with multiple case reports published in the literature.41,42 Overexpression of HMG-CoA reductase in genetically susceptible patients may explain this event.76–80

09 Diseases of the Visual System

Evidence regarding associations of statins with diseases of the visual system is heterogeneous and divergent in nature. Some studies suggest that statins have potential pleiotropic effects, whereas others show a potential risk of developing certain eye disorders, such as cataracts.44,45,81 A recent SR assessed the beneficial effects of statins on ocular disorders, concluding that statins appear to have a protective role in cataract formation, age-related macular degeneration, diabetic retinopathy and non-infectious uveitis.82 However, the authors noted that the available data are insufficient to enable a robust conclusion to be reached.82

13 Diseases of the Digestive System

Wang et al. identified an increased risk of developing gastrointestinal discomfort related to statins.18 The results of the studies included in that SR seem to corroborate an increased risk of liver injury or hepatotoxicity with statin therapy. The spectrum of liver injury associated with statins ranges from asymptomatic, transient elevations of hepatic transaminases to very rare, severe idiosyncratic liver failure.83,84 The mechanisms underlying these events are not fully understood but may involve genetic predisposition, mitochondrial dysfunction, oxidative stress and immune-mediated responses.83,84 Despite this evidence, there are also studies reporting potential hepatoprotective and therapeutic effects associated with statin use.85

15 Diseases of the Musculoskeletal System or Connective Tissue

Some of the SRs reported a statistically significant small risk of musculoskeletal events, with RRs ranging from 1.03 to 1.13.18,32,49 However, these effects appear to have limited clinical relevance. Limitations identified in the SRs were heterogeneity, lack of studies or the rarity of certain event types. Higher-intensity statin therapy was associated with a small but significantly higher incidence of patient-reported muscle complaints. The evidence suggests that these symptoms could be dose-dependent and related to systemic exposure to active statin components.6 The aetiology of these symptoms is complex. Several hypotheses have been proposed, including mitochondrial dysfunction, genetic predisposition, membrane alterations, apoptosis and oxidative stress and immune-mediated mechanisms.86

In a recent study, Weninger et al. demonstrated that simvastatin can bind to the skeletal muscle ryanodine RyR1 receptor and, in the presence of RyR1 mutations, is associated with leaky RyR1 channels and muscle weakness in preclinical models.87 This mechanism may contribute to statin-associated muscle symptoms and suggests that individuals with RyR1 mutations could represent a potentially high-risk group for statin intolerance.87 This provides a biologically plausible explanation for muscle-related adverse events, helping to reconcile some of the variability observed across clinical studies and supporting the role of individual susceptibility. Despite this, current research still recommends statin therapy due to its efficacy and the low incidence of these events.6

16 Diseases of the Genitourinary System

The impact of statin therapy on renal function is complex, with evidence suggesting both protective effects and the potential for renal insufficiency.18,32,88,89 One SR found an increased risk of renal insufficiency related to statin therapy for primary prevention of cardiovascular events, particularly with rosuvastatin.32 Nonetheless, the diagnosis and measurement of this event varied among studies, which can affect the estimate of risk.32 Some SRs assessed the benefits of statin therapy on renal function; however, they concluded that large studies should be considered to evaluate these effects.88,89

18 Pregnancy, Childbirth or the Puerperium and 20 Developmental Anomalies

All included SRs estimated an increased risk of spontaneous abortion related to statin therapy. The use of statins was primarily contraindicated during pregnancy, mostly due to the crucial role of cholesterol in foetal development.90 Cholesterol is a precursor for steroid hormones, cell membranes and bile acids, all important for normal foetal development.53 Because statin therapy inhibits the synthesis of cholesterol, there were concerns about potential teratogenicity. However, the evidence is not suggestive of teratogenic effects.90 In fact, the Food and Drug Administration has removed the contraindication of statin therapy during pregnancy.91 Nonetheless, continuous monitoring of patients taking statins during pregnancy (e.g. those with familial hypercholesterolaemia) is recommended, and the use ofthese drugs should not be deliberated.91,92

21 Symptoms, Signs or Clinical Findings, Not Elsewhere Classified

Statins appear to elevate transaminases, particularly atorvastatin, for which consistent associations were observed across studies, and to a lesser extent lovastatin and rosuvastatin. Nevertheless, other analyses failed to confirm this relationship, highlighting the variability in the evidence.83,84 Elevations of hepatic transaminases are generally asymptomatic and transient and, typically, less than three times the upper limit of normal.82,83 These elevations are often self-limiting, frequently resolving spontaneously even with continued statin therapy and are generally not indicative of a significant liver injury.82,83 Therefore, current clinical guidelines do not recommend routine monitoring of hepatic enzymes, but rather baseline assessment and subsequent monitoring only if clinical symptoms appear.6

A recent publication by Reith et al. not included in this umbrella review due to its publication outside the predefined time frame, provides further relevant evidence.93 That analysis identified a significant association between statin therapy and elevations in liver transaminases and other liver function test abnormalities, particularly with higher-intensity regimens and atorvastatin, broadly consistent with the findings of the present review.93 However, the study found no evidence of a causal relationship for most other outcomes listed as potential AEs on statin labels.93 These findings are largely in line with the present umbrella review, which also found no clear causal association for outcomes with conflicting evidence, such as renal injury, thereby reinforcing the overall favourable safety profile of statins.

Individual Statins

Among individual statins, rosuvastatin and atorvastatin stood out for their significantly increased risk of AEs. Rosuvastatin was consistently associated with NOD and showed associations with elevated transaminases, renal insufficiency, glaucoma and other eye conditions not otherwise specified.31–33,46,61,93 Rosuvastatin was associated with a slightly increased risk of self-reported muscle symptoms.32 Atorvastatin was also consistently linked to NOD.31,33,58,93 Other studies demonstrated that atorvastatin could be linked to liver injury and transaminase elevation.18,32,61 Fluvastatin was linked to a higher risk of NOD in one SR and showed a potential for a slightly increased risk of liver injury compared with other statins, although this observation was inconsistent across the studies.31,32,48 Lovastatin appears to have a neutral effect on NOD, but was associated with an increased risk of liver injury in two MAs and had conflicting evidence regarding cancer risk.18,24,25,32–34,61 Simvastatin was not associated with liver injury or transaminase elevation, but was linked to a higher risk of skin cancer and glaucoma, while evidence for its association with NOD remains inconsistent.18,27,31–33,46,48,61 Pravastatin was associated with a higher risk of glaucoma in one study, whereas most studies showed no significant association with NOD or with liver injury and transaminase elevation.18,31–34,46,48,61 In contrast, pitavastatin was the only statin to demonstrate a potential protective effect, specifically by reducing the risk of NOD.32,35 Pitavastatin was also found to be associated with a lower risk of NOD than atorvastatin and rosuvastatin.32,35 Pitavastatin showed no consistent association with liver injury or transaminase elevation.18,32,49

This umbrella review highlights notable differences in the safety profiles of individual statins, underscoring the importance of selecting therapy according to patient characteristics and risk factors. Although the overall safety of statins remains well established, the evidence reveals class- and molecule-specific variations in the risk of certain AEs. Notably, the risk of metabolic and hepatic AEs appears particularly pronounced with rosuvastatin and atorvastatin, whereas pitavastatin demonstrates a more favourable safety profile. The safety profile of statins can be influenced by several factors, such as dose intensity, hydrophilic versus lipophilic profile, individual patients, including the presence of genetic variations and certain pre-existing conditions, and potential drug–drug interactions.6 Hydrophilic statins are related to a more favourable safety profile regarding glucose metabolism than lipophilic statins, which can have extrahepatic effects by penetrating muscle and fat cells, interfering with the mevalonate pathway in peripheral tissues, which is one of the proposed mechanisms for the development of NOD.69,70,94 Some studies showed that genetic variations could influence the pharmacokinetic profile, namely by increasing area under the plasma concentration–time curve and consequently increasing the concentration of the drug in the plasma.76,95 One well-established example is the risk of statin-induced myopathy.76,95 Another difference in the pharmacokinetic profiles of statins includes the metabolism of atorvastatin, lovastatin and simvastatin by CYP3A4, which makes them more susceptible to drug–drug interactions.96 Overall, clinical decision-making should balance the potent lipid-lowering efficacy of statins with their differential safety profiles, reinforcing the need for an individualised approach to statin therapy to optimise patient outcomes.

Strengths and Limitations

This umbrella review synthesises the results of SRs and/or MAs published over the past decade, providing a high-level overview of the cumulative evidence regarding the safety of statin therapy. The review also provides a comprehensive assessment of varied safety outcomes and helps identify knowledge gaps requiring further research.

The results of this umbrella review should be interpreted carefully, considering the heterogeneity among the included SRs and/or MAs, such as differences in demographics, methods and risk-of-bias scores. The umbrella review included a comprehensive search in the two major bibliographic databases of the biomedical literature. However, some SRs and/or MAs may not have been included if they were not indexed in these databases. Only SRs and/or MAs published in the past 10 years were included. The main reason for this is that most of the included SRs performed searches from the time of database inception, meaning that all previous literature was reviewed. Nonetheless, we may have missed relevant studies. It is also of note that the results of the methodological quality assessment showed that approximately 86% of the included SRs and/or MAs were rated as having Critically Low overall confidence. However, this was due primarily to the non-reporting of a protocol, a list of excluded studies, the reasons for exclusion and sources of funding for each included study.

Conclusion

This umbrella review highlights the complex and often conflicting nature of evidence regarding statin safety. Although statins remain broadly safe and effective, concerns persist regarding NOD, musculoskeletal symptoms and liver injury/dysfunction or transaminase elevation. The evidence points to an increased risk of NOD, particularly with rosuvastatin and atorvastatin, in an intensity-dependent manner. Conversely, pitavastatin appears to be the only statin with a potential protective effect, particularly regarding NOD. Liver injury/dysfunction and elevated transaminases were observed, along with a small risk of muscle-related symptoms, specifically at higher statin intensity. All studies evaluating pregnancy outcomes reported an increased risk of spontaneous abortion with statin use. Statin-associated necrotising myopathy is a rare but serious AE, with case reports published in the literature. Statin use may also be linked to a slightly increased risk of influenza, herpes zoster and cataracts, although the results remain inconsistent and are limited by high heterogeneity. Overall, these findings reinforce the favourable safety profile of statins, but differences between individual statins and intensities should be integrated into clinical decision-making, and further research is needed to clarify these safety outcomes.

Clinical Perspective

• Statins have distinct safety profiles, requiring selection based on individual patient characteristics.

• Rosuvastatin and atorvastatin are most frequently associated with endocrine and metabolic adverse outcomes, whereas pitavastatin is associated with a potential reduction in the risk of new-onset diabetes.

• Higher statin intensity was associated with liver dysfunction, transaminase elevations and a small increase in muscle-related symptoms.

• All statins studied carry a risk of spontaneous abortion.

Funding Statement

This work was supported by Tecnimede - Sociedade Técnico-medicinal SA.