Introduction
Atherosclerotic cardiovascular disease (ASCVD) remains a leading cause of mortality worldwide, with hypercholesterolemia—particularly elevated low-density lipoprotein cholesterol (LDL-C)—established as a causal and modifiable risk factor (Borén et al., 2020; GBD 2019 Diseases and Injuries Collaborators, 2020). Statins, which inhibit cholesterol synthesis by targeting 3-hydroxy-3-methyl glutaryl coenzyme A reductase, have long been the cornerstone of lipid-lowering therapy and have demonstrated significant reductions in cardiovascular events in both primary and secondary prevention settings (Silverman et al., 2016; Yebyo et al., 2019).
However, real-world evidence consistently demonstrates that many patients, particularly those with high baseline LDL-C or established ASCVD, fail to achieve guideline-recommended LDL-C targets despite maximum tolerated statin doses (Gitt et al., 2016; Ma et al., 2021). The EUROASPIRE IV and DYSIS studies reported that 42–60% of high-risk patients remained above their target LDL-C levels with statin therapy alone (Gitt et al., 2012; Reiner et al., 2016). This treatment gap has become more pronounced as clinical guidelines have progressively established more stringent LDL-C goals for high-risk populations; thus, the combination of statins with nonstatin lipid-lowering agents has been recommended (Blumenthal et al., 2026; Joint Committee on the Chinese Guidelines for Lipid Management et al., 2023; Mach et al., 2020; Mach et al., 2025).
Ezetimibe, the most commonly used nonstatin, selectively inhibits intestinal cholesterol absorption by targeting the Niemann-Pick C1-Like 1 protein, complementing the inhibition of hepatic cholesterol synthesis by statins (Kosoglou et al., 2005; Olmastroni et al., 2024). This dual-mechanism approach addresses the compensatory increase in cholesterol absorption often observed with statin therapy alone. The landmark IMPROVE-IT trial demonstrated that adding ezetimibe to simvastatin resulted in a 24% further reduction in LDL-C levels and reduced the risk of cardiovascular events by 6.4% (hazard ratio (95% confidence interval (CI)): 0.936 [0.89–0.99]) compared to simvastatin alone, establishing the clinical benefit of combination therapy (Cannon et al., 2015).
Currently, the combined statin and ezetimibe therapy is recommended as a primary strategy in major guidelines, particularly for very high-risk patients with persistent hypercholesterolemia despite maximally tolerated statins (Blumenthal et al., 2026; Grundy et al., 2019; Jacobson et al., 2015; Mach et al., 2020; Mach et al., 2025).
Substantial evidence supports the use of fixed-dose combination (FDC) strategies as effective interventions to enhance medication adherence in chronic diseases, such as hypertension (Bangalore et al., 2007; Castellano et al., 2014; Gupta, Arshad & Poulter, 2010). Translating this approach to dyslipidemia management, FDCs combining ezetimibe with atorvastatin—a high-potency statin with extensive clinical validation—have been developed to address persistent challenges in achieving optimal LDL-C targets (Ferreira & Marques da Silva, 2017). Although multiple studies have compared atorvastatin monotherapy to atorvastatin/ezetimibe (AS/EZ) combinations (both free and fixed-dose), comprehensive comparative evidence synthesizing all three treatment strategies is lacking (Czech et al., 2022; Deharo et al., 2014).
This evidence gap contributes to the underutilization of combination strategies in clinical practice despite their potential benefits.
The present systematic review and network meta-analysis aimed to comprehensively evaluate the comparative effectiveness and safety of atorvastatin monotherapy, AS/EZ free-equivalent combination (FEC), and AS/EZ FDC in patients with hypercholesterolemia. By integrating direct and indirect evidence across randomized controlled trials, we sought to provide robust evidence to inform clinical decision-making for optimal lipid management.
Materials and Methods
Search strategy and selection criteria
This systematic review and network meta-analysis were conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines for network meta-analyses (Hutton et al., 2015). The protocol was registered prospectively on PROSPERO (CRD42024625701).
We systematically searched PubMed, Embase, the Cochrane Library, Web of Science, and Clinicaltrials.gov from inception to March 2025. The search strategy combined the medical subject heading terms “drug combinations,” “atorvastatin,” and “ezetimibe,” along with associated entry terms. The search strategy for each database is presented in Table S1.
Studies were eligible if the following conditions were met: (1) adult patients with hypercholesterolemia were enrolled; (2) at least two of the following interventions were compared: atorvastatin monotherapy, AS/EZ FEC, or AS/EZ FDC; (3) the percentage change from baseline of LDL-C levels, LDL-C goal attainment rates, or safety outcomes was reported; (4) the treatment duration was at least 4 weeks; and (5) they were randomized controlled trials (RCTs). We excluded studies with single-dose administration (i.e., pharmacokinetic studies) or short-term treatment durations (<4 weeks), those not reporting outcomes of interest, review articles, comments, letters, and meeting abstracts.
Data extraction and quality assessment
Two investigators (Xingxin Zhang and Sisi Chen) independently screened titles, abstracts, and full texts of identified studies. Data extraction was performed using a standardized form that captured: (1) study characteristics (author, publication year, treatment regimen, and sample size); (2) patient characteristics (age, sex, statin use at baseline, and ASCVD status); (3) intervention details (dosage, frequency, and duration); and (4) outcomes (mean and standard deviations of LDL-C changes and numbers of patients with LDL-C goal attainment and adverse events [AEs]).
Two investigators (Xingxin Zhang and Sisi Chen) independently assessed study quality by using the Cochrane Risk of Bias 2 tool, which evaluates six domains: randomization process, deviations from intended interventions, missing outcome data, measurement of the outcome, selection of the reported results, and overall bias (Sterne et al., 2019). Disagreements in screening, extraction, or quality assessment were resolved through discussion or consultation with a third investigator (Juying Qian).
Outcomes
The primary outcomes were as follows: (1) percentage change from baseline in LDL-C levels and (2) LDL-C goal attainment rates according to applicable guidelines (Blumenthal et al., 2026; Mach et al., 2025). Secondary outcomes included safety profiles, focusing on adverse drug reactions (ADRs; adverse reactions considered related to the study drugs in the original trials), discontinuations due to AEs, over all AEs (defined as the total number of patients experiencing any AE as reported in the original trials) and two specific AEs of clinical relevance to statin/ezetimibe therapy: muscle-related symptoms and liver function abnormality (indicated by transaminase elevation).
Statistical analysis
We performed a network meta-analysis using a hierarchical random-effects model to synthesize direct and indirect evidence (Ferreira & Marques da Silva, 2017). The network included all three interventions: atorvastatin monotherapy, AS/EZ FEC, and AS/EZ FDC. The analysis was conducted using Markov Chain Monte Carlo simulations with 50,000 iterations (including 10,000 burn-in iterations).
Network consistency was assessed using loop-specific inconsistency analysis and the node-splitting approach (Czech et al., 2022). We calculated the surface under the cumulative ranking curve (SUCRA) to rank treatments, with higher SUCRA values (closer to 100%) indicating greater probability of being the best treatment.
For the percentage change from baseline in LDL-C levels, we calculated the mean differences with 95% CIs using a random-effects model. For LDL-C goal attainment and safety outcomes, we calculated the risk ratios (RRs) with 95% CIs. Heterogeneity was assessed using the I2 statistic and Cochrane’s Q test, with I2 values of 25%, 50%, and 75% considered low, moderate, and high heterogeneity, respectively.
Pairwise meta-analyses were conducted using R software (version 4.2.0; R Core Team, 2022) with the metafor and meta packages. Network meta-analysis analyses were performed using R with the gemtc package and the Network package in STATA (Version 14.0; StataCorp LLC, College Station, TX, USA). Results with p < 0.05 were considered statistically significant.
Results
Search results and study characteristics
The literature search identified 3,327 records, with 71 studies meeting the inclusion criteria for the systematic review. After excluding studies with insufficient or duplicate data (Table S2), 49 independent datasets from 46 trials with 10,567 participants were included in the network meta-analysis (Anusha et al., 2023; Arimura et al., 2012; Azar et al., 2011; Azar et al., 2010; Ballantyne et al., 2003b; Ballantyne et al., 2004; Barbosa et al., 2013; Bays et al., 2013; Bays et al., 2015; Blagden & Chipperfield, 2007; Conard et al., 2008; Constance et al., 2014; Cruz-Fernández et al., 2005; El-Tamalawy et al., 2018; Her et al., 2010; Jackowska et al., 2016; Japaridze & Sadunishvili, 2017; Kanat et al., 2009; Lee et al., 2023; Lee et al., 2011; Lee et al., 2012; Leiter et al., 2008; Liu, Liu & Xiao, 2018; Liu et al., 2017; Luo et al., 2014; Luo et al., 2016; Malina et al., 2015; Matsue et al., 2013; Nicholls et al., 2017; Oh et al., 2021; Ostad et al., 2009; Padhy, Yadav & Gupta, 2013; Philip & Elizabeth, 2021; Piorkowski et al., 2007; Pytel et al., 2017; Pytel et al., 2016; Qian et al., 2022; Stein et al., 2004; Tan et al., 2021; Teramoto et al., 2018; Teramoto et al., 2012; Tsujita et al., 2015; Ueda et al., 2017; Uemura et al., 2012; Wu et al., 2018; Zieve et al., 2010). Three studies each reported two individual datasets (Bays et al., 2015; Nicholls et al., 2017; Qian et al., 2022).
The PRISMA flowchart for study selection is shown in Fig. 1.
The included trials were published between 2003 and 2023, with sample sizes ranging from 18 to 1,053 participants. Eleven (23.9%) studies enrolled statin-naive patients, while 35 (76.1%) studies enrolled statin-treated patients. Forty-three (93.5%) studies investigated patients with ASCVD.
Treatment duration varied from 6 weeks to 12 months. The study population consisted of adults with hypercholesterolemia (mean age, 55–64 years), with the baseline LDL-C levels ranging from 3.1 to 4.9 mmol/L. Across studies, atorvastatin monotherapy was administered at doses ranging from 10 to 80 mg/day, with one study using 30 mg/day and one study using dynamic dose adjustment.
In the AS/EZ combination arms, the atorvastatin component also varied, whereas ezetimibe was generally administered at 10 mg/day. Some trials used dose-matched comparisons between atorvastatin monotherapy and AS/EZ combination therapy, while others compared AS/EZ combination therapy with higher-dose atorvastatin monotherapy. The details of the included studies are summarized in Table 1.
Risk of bias assessment
The overall methodological quality of the included trials was low-to-some concern. Forty-five studies had a low risk of bias, and only one study had some concern of risk of bias (Table S3).
Primary outcome: percentage change in LDL-C
Network meta-analysis
The percentage change in LDL-C was reported in 46 datasets from 43 RCTs, involving 8,751 patients. The network structure for the percentage change in LDL-C is illustrated in Fig. 2A, which shows the direct comparisons available between treatments, with the node size corresponding to the sample size of the treatment group and the edge thickness indicating the number of direct comparisons between treatments.
The network meta-analysis corroborated the findings from the pairwise comparisons, estimating mean differences of 12.41% (95% CI [10.22–14.61]%) for AS/EZ FEC vs. atorvastatin monotherapy and 13.94% (95% CI [9.10–18.78]%) for AS/EZ FDC vs. atorvastatin monotherapy. Additionally, AS/EZ FDC showed no clinically meaningful advantage over AS/EZ FEC in LDL-C reduction (difference: 1.53%, 95% CI [−3.49 to 6.55%]) (Figs. 2B–2C).
Treatment ranking based on the SUCRA values indicated that AS/EZ FDC had the highest probability (86.2%) of being the most effective treatment for LDL-C reduction, followed by AS/EZ FEC (63.8%). In comparison, atorvastatin monotherapy ranked the lowest (SUCRA of ∼0%), as visualized in Fig. 2D.
Loop-specific inconsistency analysis showed no significant inconsistency between direct and indirect evidence (all p > 0.05), supporting the validity of the network model.
Pairwise meta-analysis
Forest plots of direct pairwise comparisons detailing the percentage change in LDL-C by treatment group are presented in Fig. S1. The pooled mean LDL-C percentage reduction was −12.32% (95% CI [−14.69% to −9.96%]) (I2 = 91%) for AS/EZ FEC vs. atorvastatin monotherapy and −14.43% (95% CI [−19.03% to −9.82%]) (I2 = 86%) for AS/EZ FDC vs. atorvastatin monotherapy.
The pooled mean LDL-C percentage reduction was comparable between AS/EZ FEC and AS/EZ FDC (−0.20%, 95% CI [−0.49% to 0.09%]; I2 = 0%).
Subgroup analysis
Subgroup analyses according to prior statin treatment status were further performed for the percentage change in LDL-C. In statin-treated patients, the network meta-analysis of 36 datasets from 33 studies showed greater LDL-C reduction with both combination regimens than with atorvastatin monotherapy, with mean differences of 13.90% (95% CI [11.47–16.33]%) for AS/EZ FEC vs. atorvastatin monotherapy and 14.41% (95% CI [9.81–19.00]%) for AS/EZ FDC vs. atorvastatin monotherapy. AS/EZ FDC showed no clinically meaningful advantage over AS/EZ FEC (difference: 0.50%, 95% CI [−4.32 to 5.33%]) (Fig.
S2). Treatment ranking in this subgroup indicated that AS/EZ FDC had the highest probability of being the most effective treatment for LDL-C reduction (SUCRA: 79.0%), followed by AS/EZ FEC (71.0%), whereas atorvastatin monotherapy ranked lowest (SUCRA: 0.0%). The direct pairwise comparisons in statin-treated patients showed pooled mean LDL-C percentage reductions of −13.88% (95% CI [−16.54% to −11.21%]; I2 = 91.9%) for AS/EZ FEC vs. atorvastatin monotherapy, −14.43% (95% CI [−19.03% to −9.82%]; I2 = 85.9%) for AS/EZ FDC vs. atorvastatin monotherapy, and −0.20% (95% CI [−0.49% to 0.09%]; I2 = 0.0%) for AS/EZ FDC vs.
AS/EZ FEC (Fig. S3). In statin-naïve patients, only direct evidence for AS/EZ FEC vs. atorvastatin monotherapy was available, showing a pooled mean LDL-C percentage reduction of −7.47% (95% CI [−12.01% to −2.93%]; I2 = 81.4%) (Fig.
S4).
Primary outcome: LDL-C goal attainment
Network meta-analysis
LDL-C goal attainment was reported in 19 datasets from 19 studies involving 5,246 patients. The network plot for LDL-C goal attainment is shown in Fig. 3A, with no loop formed due to a lack of direct comparison between AS/EZ FEC and AS/EZ FDC.
Network meta-analysis indicated that the RR was 1.87 (95% CI [1.47–2.39]) for AS/EZ FEC vs. atorvastatin monotherapy and 1.80 (95% CI [0.85–3.81]) for AS/EZ FDC vs. atorvastatin monotherapy (Figs. 3B–3C). AS/EZ FEC had the highest probability (77.0%) of being the most effective treatment for LDL-C goal attainment, followed by AS/EZ FDC (69.9%), while atorvastatin monotherapy ranked the lowest (SUCRA of 3.2%) (Fig.
3D).
Pairwise meta-analysis
For the direct AS/EZ FEC vs. atorvastatin monotherapy comparison, a random-effects model yielded an RR of 1.87 (95% CI [1.46–2.40]) with high heterogeneity (I2 = 95%). For the direct AS/EZ FDC vs. atorvastatin monotherapy comparison, the pooled RR was 1.93 (95% CI [1.27–2.94]) with low heterogeneity (I2 = 32%) (Fig. S5).
Subgroup analysis
Subgroup analyses were also performed for LDL-C goal attainment according to prior statin treatment status. In statin-treated patients, network meta-analysis of 14 datasets from 14 studies indicated that the RR was 2.02 (95% CI [1.54–2.65]) for AS/EZ FEC vs. atorvastatin monotherapy and 1.80 (95% CI [0.89–3.64]) for AS/EZ FDC vs. atorvastatin monotherapy. AS/EZ FDC showed no clear advantage over AS/EZ FEC (RR: 0.89, 95% CI [0.42–1.89]) (Fig.
S6). Treatment ranking suggested that AS/EZ FEC had the highest probability of being the most effective treatment for LDL-C goal attainment (SUCRA: 81.2%), followed by AS/EZ FDC (66.2%), whereas atorvastatin monotherapy ranked lowest (SUCRA: 2.6%). In direct pairwise comparisons among statin-treated patients, AS/EZ FEC was associated with a higher LDL-C goal attainment rate than atorvastatin monotherapy (RR: 2.02, 95% CI [1.53–2.67]; I2 = 92.5%), and AS/EZ FDC also increased LDL-C goal attainment compared with atorvastatin monotherapy (RR: 1.93, 95% CI [1.27–2.94]; I2 = 31.8%) (Fig.
S7). In statin-naïve patients, only the direct AS/EZ FEC vs. atorvastatin monotherapy comparison was available, with a pooled RR of 1.60 (95% CI [0.97–2.65]; I2 = 90.4%) (Fig. S8).
Secondary outcomes: safety profile
ADRs
Twenty datasets from 18 studies reported ADRs assessed by the study investigators. The network plot for ADRs is shown in Fig. 4A.
Network meta-analysis indicated that the RRs were 1.07 (95% CI [0.90–1.26]) for AS/EZ FEC vs. atorvastatin monotherapy and 1.30 (95% CI [0.94–1.81]) for AS/EZ FDC vs. atorvastatin monotherapy (Figs. 4B–4C). Atorvastatin monotherapy had the highest probability (85.6%) of no ADRs, followed by AS/EZ FEC (69.9%), while AS/EZ FDC ranked the lowest (SUCRA 8.8%) (Fig.
4D).
Direct comparisons are shown in Fig. S9. AS/EZ FEC did not increase the risk of ADR compared to atorvastatin monotherapy (RR: 1.02, 95% CI [0.86–1.21]).
For AS/EZ FDC vs. atorvastatin monotherapy, the pooled RR was 1.28 (95% CI [0.82–2.01]), and for AS/EZ FDC vs. AS/EZ FEC, it was 1.24 (95% CI [0.79–1.94]). None of these comparisons showed statistically significant differences in the occurrence of ADRs.
Discontinuation due to AEs
Twenty-six datasets from 23 studies reported discontinuation due to AEs. The network meta-analysis is shown in Fig. S10.
The RRs were 0.88 (95% CI [0.63–1.24]) for AS/EZ FEC vs. atorvastatin monotherapy and 0.79 (95% CI [0.42–1.47]) for AS/EZ FDC vs. atorvastatin monotherapy. AS/EZ FDC ranked the highest (70.7%), and atorvastatin monotherapy ranked the lowest (23.2%).
As depicted in Fig. S11, the direct pooled RRs of discontinuation due to AEs were 0.95 (95% CI [0.69–1.20]) for AS/EZ FEC vs. atorvastatin monotherapy, 1.47 (95% CI [0.57–3.80]) for AS/EZ FDC vs. atorvastatin monotherapy, and 0.62 (95% CI [0.30–1.31]) for AS/EZ FDC vs. AS/EZ FEC (Fig.
S11).
Overall AEs
Twenty-six datasets from 23 studies reported overall AEs. Network meta-analysis is shown in Fig. S12.
The RRs were 1.00 (95% CI [0.92–1.09]) for AS/EZ FEC vs. atorvastatin monotherapy and 1.05 (95% CI [0.92–1.21]) for AS/EZ FDC vs. atorvastatin monotherapy. AS/EZ FEC and atorvastatin monotherapy ranked the highest at similar levels (63.5%) and AS/EZ FDC ranked the lowest (23.0%) in terms of the numbers of AEs.
Direct comparison between groups in terms of AEs is illustrated in Fig. S13. The direct pooled RRs were 0.96 (95% CI [0.89–1.03]) for AS/EZ FEC vs. atorvastatin monotherapy, 1.11 (95% CI [0.91–1.35]) for AS/EZ FDC vs. atorvastatin monotherapy, and 1.00 (95% CI [0.84–1.91]) for AS/EZ FDC vs.
AS/EZ FEC.
Specific AEs: muscle-related symptoms and liver function abnormality
Muscle-related symptoms.
Nineteen datasets from 17 studies reported muscle-related symptoms. The network meta-analysis is shown in Fig. S14.
The RRs were 1.08 (95% CI [0.67–1.76]) for AS/EZ FEC vs. atorvastatin monotherapy and 1.00 (95% CI [0.26–3.92]) for AS/EZ FDC vs. atorvastatin monotherapy. Atorvastatin monotherapy ranked the highest (57.0%) and AS/EZ FEC ranked the lowest (41.4%) in terms of the number of muscle-related symptoms.
Direct comparison between groups in terms of muscle-related symptoms is illustrated in Fig. S15. The direct pooled RRs were 1.13 (95% CI [0.71–1.82]) for AS/EZ FEC vs. atorvastatin monotherapy, 0.68 (95% CI [0.12–4.01]) for AS/EZ FDC vs. atorvastatin monotherapy, and 3.04 (95% CI [0.12–74.37]) for AS/EZ FEC vs.
AS/EZ FDC.
Liver function abnormalities.
Twenty-one datasets from 20 studies reported liver function abnormalities. The network meta-analysis is shown in Fig. S16.
The RRs were 0.66 (95% CI [0.41–1.06]) for AS/EZ FEC vs. atorvastatin monotherapy and 0.36 (95% CI [0.10–1.31]) for AS/EZ FDC vs. atorvastatin monotherapy. AS/EZ FDC ranked the highest (88.3%) and atorvastatin monotherapy ranked the lowest (4.9%) in terms of liver function abnormalities.
Direct comparison between groups in terms of liver function abnormalities is illustrated in Fig. S17. The direct pooled RRs were 0.67 (95% CI [0.42–1.06]) for AS/EZ FEC vs. atorvastatin monotherapy, 0.28 (95% CI [0.05–1.67]) for AS/EZ FDC vs. atorvastatin monotherapy, and 0.68 (95% CI [0.11–4.08]) for AS/EZ FDC vs.
AS/EZ FEC.
Publication bias
We assessed publication bias for all outcomes. The comparison-adjusted funnel plot did not exhibit significant asymmetry for any outcomes except LDL-C goal attainment (Figs. S18–S24).
Therefore, no evidence of publication bias was found for these outcomes in the included studies. For LDL-C goal attainment, more studies reported positive results (Fig. S19).
Discussion
This systematic review and network meta-analysis comprehensively evaluated the comparative effectiveness and safety of three lipid-lowering strategies: atorvastatin monotherapy, AS/EZ FEC, and AS/EZ FDC. Our findings demonstrate that both combination approaches substantially outperform atorvastatin monotherapy in reducing LDL-C levels and achieving guideline-recommended targets, with the FDC showing the greatest effectiveness. Importantly, these efficacy advantages were achieved without significant increases in ADRs or discontinuation rates.
The magnitude of LDL-C reduction observed with AS/EZ FEC (additional 12.41% beyond monotherapy) aligns with previous studies reporting 15–20% incremental LDL-C lowering when ezetimibe is added to statins (Ballantyne et al., 2003a; Morrone et al., 2012). This effect reflects ezetimibe’s complementary mechanism of blocking intestinal cholesterol absorption while statins inhibit hepatic synthesis. AS/EZ FDC had the highest SUCRA value (86.2%), followed by the AS/EZ FEC (63.8%), indicating that AS/EZ FDC might be the most effective treatment for LDL-C reduction, which may partially be attributed to possible improved adherence.
Two meta-analyses have shown that FDC can enhance medication adherence in hypertension (Bangalore et al., 2007; Gupta, Arshad & Poulter, 2010). Real-world studies also have found that the FDC strategy significantly improved adherence (Bartlett, Pratt & Roughead, 2017; Rea et al., 2021). Moreover, it has been reported that a high adherence is related to lower LDL-C levels (Chi et al., 2014; George & Bender, 2019).
Our analysis of LDL-C goal attainment found that patients receiving combination therapy were approximately twice as likely to achieve target levels compared to those receiving monotherapy. The performance of AS/EZ FDC was similar to AS/EZ FEC (RR: 0.96), suggesting that the simplified regimen provides comparable efficacy. While the AS/EZ FDC regimen may not offer superior pharmacological efficacy, its potential to enhance patient adherence (Bartlett, Pratt & Roughead, 2017; Rea et al., 2021) could translate into improved clinical outcomes in real-world settings (Katzmann et al., 2022).
The subgroup analyses according to prior statin treatment status generally supported the robustness of the primary findings. In statin-treated patients, both AS/EZ FEC and AS/EZ FDC were associated with greater LDL-C reduction and higher LDL-C goal attainment than atorvastatin monotherapy, consistent with the overall network meta-analysis. This finding is also consistent with the mechanistic rationale that statin-induced suppression of hepatic cholesterol synthesis may be accompanied by compensatory enhancement of intestinal cholesterol absorption, thereby increasing the relevance of NPC1L1 inhibition with ezetimibe in patients already receiving statin therapy (Thongtang et al., 2012; Tremblay et al., 2011).
In statin-naïve patients, the available evidence was limited and mainly involved direct comparisons between AS/EZ FEC and atorvastatin monotherapy. Therefore, findings in this subgroup should be interpreted cautiously, and further studies are needed to determine whether the relative effects of FDC and FEC differ according to prior statin exposure.
The atorvastatin dose should be considered when interpreting the magnitude of LDL-C reduction and LDL-C goal attainment. The included RCTs did not use a uniform atorvastatin dose across all comparisons; some studies compared AS/EZ combination therapy with atorvastatin monotherapy at the same statin dose, whereas others compared combination therapy with higher-dose atorvastatin monotherapy. This design reflects real-world clinical decision-making, in which physicians may either increase the statin dose or add ezetimibe when LDL-C remains above target.
However, it also means that the pooled estimates represent average effects across different dosing strategies rather than a strictly dose-controlled comparison. The high heterogeneity observed in some direct comparisons, particularly for AS/EZ FEC vs. atorvastatin monotherapy, may be partly attributable to this variation in atorvastatin dosage.
The safety profile of combination therapy was reassuring, with no significant increase in ADRs compared to monotherapy (RR [CI]: 1.07 [0.90–1.26] and 1.30 [0.94–1.81]). The similar discontinuation rates and AEs across treatment groups (combination vs. monotherapy, RR of discontinuation rate [CI]: 0.88 [0.63–1.24] and 0.79 [0.42–1.47]; RR of AEs [CI]: 1.00 [0.92–1.09] and 1.05 [0.92–1.21]) further support the tolerability of the combination approaches, which can improve adherence (Bartlett, Pratt & Roughead, 2017; Rea et al., 2021). Moreover, for severe AEs such as muscle-related symptoms and liver function abnormalities, the AS/EZ FDC regimen demonstrated a higher SUCRA value than AS/EZ FEC, suggesting a potentially lower incidence of these AEs with the former.
These findings address an important clinical concern, as safety considerations often influence prescribing decisions for lipid-lowering therapies (Stroes et al., 2015).
Our study has several strengths. First, it represents the most comprehensive synthesis to date comparing all three treatment strategies for hypercholesterolemia. Second, the network meta-analysis approach allowed for integration of both direct and indirect evidence, enabling comparisons even where head-to-head trials were limited.
Third, the analysis was restricted to RCTs, with no significant inconsistency across the network meta-analysis. Fourth, the included studies involved patients from diverse geographic regions and countries, reflecting a broad range of ethnic and clinical backgrounds and thereby enhancing the generalizability of the findings. Finally, the inclusion of both efficacy and safety outcomes provides a balanced assessment of the risk-benefit profile for each strategy.
Nevertheless, several limitations warrant consideration. First, between-study heterogeneity was substantial for LDL-C outcomes, reflecting diversity in study populations, dosing regimens, and baseline characteristics. Second, the number of studies directly comparing FDC to FEC was limited, providing less evidence to compare when performing the network meta-analysis.
Third, our analysis focused on intermediate outcomes (LDL-C reduction and goal attainment) rather than hard clinical endpoints such as cardiovascular events or mortality. Therefore, most included trials had relatively short durations (≤12 months), limiting assessment of long-term efficacy and safety.
Conclusions
In conclusion, in patients with hypercholesterolemia, the AS/EZ FDC regimen demonstrated superior efficacy in lowering LDL-C levels and achieving guideline-recommended targets compared to atorvastatin monotherapy, while showing comparable effectiveness to AS/EZ FEC and without an increase in ADRs or AEs. Therefore, AS/EZ FDC may represent an effective and convenient lipid-lowering option, particularly when combination therapy is indicated. Further studies are warranted to determine whether the simplified fixed-dose regimen translates into improved adherence, treatment persistence, and cardiovascular benefit in clinical practice.
Supplemental Information