FormalPara Key Summary Points

Why carry out this study?

Observational studies suggest a log-linear relationship between LDL-C concentrations and CV events.

LDL-C reduction is a must to prevent atherothrombosis and plaque rupture, which portend high CV morbidity and mortality.

Real-world evidence studies demonstrate a lack of target LDL-C reductions (C-LDL < 55 mg/dL) despite the use of optimized LLT.

What was learned from this study?

In Spain, only 9.6% of patients with ASCVD achieve the target LDL-C reduction according to the 2019 ESC/EAS goals (LDL-C < 55 mg/dL).

Optimized LLT is used in 46.2% of patients with ASCVD, mostly based on high-intensity statins (33.8%) or combined use of statins and ezetimibe (11.2%).

Very few patients on optimized oral LLT achieve 2019 ESC/EAS treatment goals (17.9%), despite the possibility to accede to reimbursed use of PCSK9i (17.9% of patients showed LDL-C > 100 mg/dL).

Digital Features

This article is published with digital features, including a graphical abstract to facilitate understanding of the article. To view digital features for this article go to https://doi.org/10.6084/m9.figshare.21583503.

Introduction

Cardiovascular (CV) diseases are the leading cause of morbidity and mortality globally [1], being responsible for more than 4 million deaths yearly in Europe [2] and accounting for 28.3% of total deaths in Spain (2018 data) [3]. Modifiable CV risk factors (CVRF) can explain about 80% of risk of acute CV events [4] and it has been demonstrated that efficient control and intervention on the main modifiable risk factors has great potential in preventing CV events [5, 6].

Low-density lipoprotein cholesterol (LDL-C) is a well-established major causal modifiable risk factor for atherosclerotic cardiovascular disease (ASCVD) including acute events such as myocardial infarction (MI) and ischaemic stroke (IS) [7, 8]. Because the effect of LDL-C on the risk of ASCVD is both causal and cumulative over time, lowering the plasma LDL-C level is crucial to reduce the potential risk of suffering a CV event, especially in patients with prior history of CV events [8, 9].

According to the 2019 European Society of Cardiology (ESC)/European Atherosclerosis Society (EAS) dyslipidaemia guidelines [10], the LDL-C treatment goal for patients at very high CV risk (which includes all patients diagnosed with ASCVD) is less than 55 mg/dL or at least 50% reduction from baseline LDL-C level [10]. Reducing LDL-C as low and as quickly as possible is the best strategy for reducing the risk of new CV events [11,12,13]. It has been demonstrated that reduction of serum LDL-C levels to below 20 mg/dL was safe and was not associated with any impact on patients’ cognition [14, 15]. However, reduction of LDL-C to the levels recommended by the 2019 ESC/EAS dyslipidaemia guidelines remains a challenge in routine clinical practice. According to real-world data, only 18% of patients at very high risk of CV events achieve this goal [16], and the majority of such patients remain at an unnecessarily increased risk of developing subsequent acute CV events [16,17,18,19,20,21]. Current ESC/EAS guidelines recommend that for patients at very high risk of CV events whose LDL-C levels cannot be lowered below the recommended thresholds by maximally tolerated statin therapy alone, additional lipid-lowering therapies (LLT), such as ezetimibe or proprotein convertase subtilisin–kexin type 9 inhibitors (PCSK9i), should be considered [10].

In Spain, reimbursement of PCSK9i is regulated by the Therapeutic Positioning Report (TPR) [22], which sets its use to patients with familiar hypercholesterolaemia and/or patients with established CV disease with LDL-C level greater than 100 mg/dL despite the use of the maximum tolerated dose of a statin unless statins are contraindicated or a patient is intolerant to statins [22]. However, different studies suggest that in reality, access to PCSK9i seems to be restricted beyond the TPR recommendations [23, 24]. More than 27% of the patients with coronary heart disease treated with statins at maximum tolerated doses showed LDL-C levels greater than 100 mg/dL, thereby being potentially eligible for PCSK9i treatment in Spain [23]. However, only between 1% and 3.5% of these patients are currently receiving PCSK9i treatment [24].

The current study aimed to estimate the proportion and number of patients eligible for PCSK9i treatment according to the reimbursed populations included in the TPR, in routine clinical practice in Spain.

Methods

Study Design and Patients

We conducted a retrospective cohort study based on the IQVIA Spanish real-world electronic medical records (EMR) database. This database includes patient-level data with longitudinal tracking of anonymised patient histories from all specialties, mainly at outpatient level.

The main study cohort included all adult patients diagnosed with ASCVD, ordered by event severity (MI, UA, IS, TIA and/or PAD) and with event-related registered data in the database between 1 January 2014 and 31 December 2020 (n = 19,612). To be considered eligible, patients must have had at least one LDL-C or total cholesterol measurement registered during the study period and data registered at least 12 months prior to the index date, conforming to the main study cohort for socio-demographic and clinical profile description purposes (n = 9516). In addition, for primary outcome assessments (eligibility to PCSK9i), a minimum period of 3 months post-event was considered necessary to achieve LLT optimisation. Patients with at least one LDL-C or total cholesterol measurement registered after 3 months or more of the first registered event were considered eligible (n = 8739).

The index date for analysis purposes was defined as the date of the first CV event registered during the study period. The baseline period included all data registered during the year prior to the index date. The follow-up period included all data from the index date up to the last available data entry during the study period (2014–2020).

Data Management and Extraction

The IQVIA Spanish EMR database collected stable information (recurrent patients and prescribers). This information was collected from about one million patients (ca. 3.2% of overall treated population) of 8000 office-based primary and secondary care physicians, involving different healthcare centres from three Spanish regions (the regions cannot be disclosed because of contract confidentiality). The information collected was extrapolated to the national level based on demographic attributes. IQVIA assessed and confirmed the data through internal data quality check to be representative of the total Spanish population (age and gender) according to data provided by the Instituto Nacional de Estadística (INE, Spain) [3].

The database included coded variables for both patient and episode, directly extracted from the clinical practice healthcare registries in Spain, including International Classification of Diseases (ICD) for disease diagnosis or disease-related registries (e.g. events, patient visits, tests), Anatomical Therapeutic Chemical (ATC) and national code for treatment-related registries (see supplementary material). Standard patient data was collected according to the routine clinical practice in Spain: age, sex, weight, height, smoking status. Data extraction was conducted considering all adult patients with at least one registered ASCVD event during the study period.

Statistical Analysis

The analysis considered all lipid-modifying agents, classified according to ATC codes (C10). LLT intensity was defined according to the classification used by the Spanish Cardiology Society (SEC, Sociedad Española de Cardiología) [25]. Maximum tolerated dose of statins was defined as the dose received by the patient before adding a combination or the dose used at the time when the combination was initiated (defined as at least two different LLT ATC codes or based on the national code to identify fixed dose combinations). Patients on ezetimibe monotherapy, those who never used statins alone or in combinations, or those who switched from a statin to ezetimibe monotherapy were considered statin intolerant or those for whom statins were contraindicated.

Three months was assumed to be necessary for an LLT to develop its full effect on LDL-C reduction. Therefore, the analysis to assess PCSK9i eligibility only considered participants who had at least one LDL-C measurement available at least 3 months after the index date.

LLT optimization was defined according to the guideline provided by the Spanish TPR [22], as well as by the 2019 ESC/EAS guidelines [10]. For this study we considered as optimized LLT the use of high-intensity statins, a combination of a statin with ezetimibe, or use of ezetimibe monotherapy (the use of ezetimibe monotherapy was assumed to occur in statin-intolerant patients or in those for whom statins were contraindicated) and PCSK9i.

Although, for primary outcome analysis, the TPR target LDL-C threshold of LDL-C of at least 100 mg/dL was considered for PCSK9i eligibility purposes, additional LDL-C thresholds were analysed to assess the quality of dyslipidaemia control in routine clinical practice. We included two additional scenarios: (i) LDL-C treatment goal recommended by the 2019 ESC/EAS guideline (LDL-C greater than 55 mg/dL) [10] and (ii) LDL-C treatment goal recommended by the 2016 ESC guideline (LDL-C greater than 70 mg/dL) [26].

The primary analysis was performed at a single time point after database closure. All variables were described using frequency and proportion with 95% CI for categorical variables, and mean (with 95% CI), median, standard deviation (SD), interquartile range, minimum and maximum for continuous variables. In both cases, the number and proportion of missing data were reported. The comparison of use of LLT by type of patient with ASCVD was made by chi-square test.

Ethical Considerations

The study was conducted in accordance with the ethical principles of the Declaration of Helsinki and the guidelines specified in Order SAS/3470/2009 of the Agencia Española de Medicamentos y Productos Sanitarios (AEMPS) for the development of studies based on human subject data. The study was approved by the Ethics Committee for Research with Medicines of the Hospital Arnau de Vilanova (Valencia, Spain), which also approved the study to be exempt from obtaining participants’ informed consent.

Data processing complies with the Organic Law on Data Protection (Ley Orgánica 3/2018, de 5 de diciembre, de Protección de Datos Personales y garantía de los derechos digitales) as well as the General Data Protection Regulation (RGPD 2016/679).

Results

Sample Characteristics

The main study cohort included 9516 patients with ASCVD and at least one available LDL-C or total cholesterol measurement during the study period (2014–2020), corresponding to 48.5% of the total patients with ASCVD registered in the database.

Patients in the study population were on average 67.7 (SD 12.5) years of age, 63.9% were male, baseline mean (SD) LDL-C value registered was 117.3 (38.8) mg/dL and the last available LDL-C value achieved during the study period was 95.8 (35.9) mg/dL (Table 1). The median time registered from baseline to the last available LDL-C measurement was 853 days (2.3 years). About 41.3% were obese (body mass index at least 30 kg/m2), 7.2% had alcohol consumption and 3.9% smoking habit (based on categorised data; Table 1).

Table 1 Socio-demographic and clinical characteristics of the study population (based on structured variables included in the electronic medical records)
Full size table

MI was identified as the most severe event for 28.5% of patients with ASCVD overall (among the patients treated with PCSK9i this proportion reached 67.6%; Table 1). Major comorbidities related to ASCVD were diagnosed and registered in the electronic medical record for 66.7% of patients; most common among them were hypertension (47.0%), hypercholesterolaemia (34.3%) and type 2 diabetes (T2D; 22.5%). Mean (SD) Charlson comorbidity index (CCI) score was 3.1 (1.7) and 19.7% of patients had severe comorbidity diseases (CCI grade 5 or higher).

LLT Management

At any point during the study period 84.3% of patients received LLT. Among those receiving LLT, 98.5% received statins; ezetimibe (monotherapy or in combination) was used in 17.4% of patients, and 0.9% received a PCSK9i (hospital medication was fully registered in the Spanish database from 2018 onwards).

Among patients with ASCVD and at least one LDL-C measurement available 3 months or more after the index date (n = 8739), 64.4% of patients were treated with an LLT in the 3 months prior to their last available LDL-C measurement, registered in a median time of 937 days (2.6 years). Statins were used in 97.4% of the patients, ezetimibe (monotherapy or in combination) in 12.0% and PCSK9i were used in 0.4% of these patients (Table 2).

Table 2 Lipid-lowering therapy used in patients with ASCVD and at least one LDL-C measurement available in a time period at least 3 months after index date
Full size table

High-intensity statins were used in 33.8% of the patients; this proportion increased to 44.7% and 40.0% in patients with MI and UA, respectively, being lower in IS (36.7%), TIA (25.4%) and PAD (20.5%) patients, p < 0.001 (Table 2). Atorvastatin 40–80 mg (29.1%) and rosuvastatin 20–40 mg (4.8%) were the most frequently registered high-intensity statins, prior to the last available LDL-C measurement. Combinations based on statin and ezetimibe were used in 11.2% of the patients, ranging from 16.6% in patients whose most severe CV event was MI to 5.3% in those whose most severe event was IS (p < 0.001; Table 2).

In the study population 46.2% of patients received optimized LLT according to the definition included in Sect. “Statistical Analysis” (45.8% received optimized oral LLT). This proportion ranged from 30.7% in patients with PAD up to 62.8% in patients with MI. Most patients on optimized LLT received high-intensity statins (Table 2). PCSK9i were registered in 0.4% of the study population (0.6% and 0.8% for patients with MI and UA, respectively; Table 2).

LDL-C Goals Attainment

Overall, average LDL-C reduction from baseline to the last available LDL-C measurement was around 18.3%. The mean (SD) LDL-C value reached at the end of the follow-up period was 95.7 (35.8) mg/dL for the primary study cohort and in a median time point from index date of 937 days (2.6 years). The LDL-C treatment goal recommended by the 2019 ESC/EAS guidelines (LDL-C below 55 mg/dL) was achieved by 11.0% of patients using LLT. In the study population 27.6% of the treated population achieved LDL-C levels below 70 mg/dL, the treatment goal recommended by 2016 ESC/EAS guidelines that were in effect for a large portion of the study period. LDL-C goal attainment was substantially higher among patients treated with PCSK9i, despite much higher baseline LDL-C levels: 23.5% of them achieved LDL-C below 55 mg/dL and 39.7% reached LDL-C below 70 mg/dL (Fig. 1).

Fig. 1
figure 1

LDL-C level achieved according to the last LDL-C measurement available, for patients with ASCVD and at least one LDL-C measurement available at least 3 months post-event; a 2019 ESC/EAS goals, b 2016 ESC goals

Full size image

The percentage of patients achieving LDL-C treatment goals was different depending on the type of CV event. The highest proportions of patients who achieved LDL-C treatment goals were observed among patients with MI and UA. Among the patients with MI 13.7% achieved LDL-C levels below 55 mg/dL and 34.2% reached LDL-C levels below 70 mg/dL; among the patients with UA the respective proportions were 10.9% and 29.8%. The lowest proportions of those who achieved LDL-C treatment goals recommended by the 2019 and 2016 ESC/EAS guidelines were observed among patients with TIA. Only 6.4% of them achieved LDL-C levels below 55 mg/dL and 17.3% achieved LDL-C levels below 70 mg/dL (Fig. 2).

Fig. 2
figure 2

LDL-C level achieved according to the last LDL-C measurement available and most severe event, for patients with ASCVD and at least one LDL-C measurement available at least 3 months post-event; a 2019 ESC/EAS goals, b 2016 ESC goals

Full size image

Patients receiving more intensive LLT tended to achieve LDL-C treatment goals more often (Fig. 3). However, even among patients on optimized oral LLT, 82.1% had LDL-C levels exceeding 55 mg/dL, 58.4% had LDL-C levels higher than 70 mg/dL and 17.9% of patients had LDL-C levels over 100 mg/dL.

Fig. 3
figure 3

Patients with at least one LDL-C measurement available at least 3 months post-event considered at goal according to different LDL-C thresholds, by LLT intensity

Full size image

LLT tended to demonstrate their effect quite quickly: LDL-C goal attainment proportions increased during the first 4–6 weeks after the index data and then remained virtually the same with very slight downward trend throughout the remainder of the follow-up period (Fig. 4).

Fig. 4
figure 4

Patients with at least one LDL-C measurement available, considered at goal according to different LDL-C thresholds, by time period from index date

Full size image

Discussion

Overall 84.3% of patients with ASCVD received LLT at any moment during the study period. Statins were the most common drug used for lipid-lowering (98.5%), followed by ezetimibe alone or in combination with a statin (17.4%). PCSK9 inhibitors were used by less than 1% of patients. Among the patients with ASCVD who received LLT, only 45.8% received an optimized oral regimen (high-intensity statins, a combination of statins and ezetimibe or ezetimibe monotherapy) and 17.9% had their LDL-C levels above 100 mg/dL despite receiving optimized oral LLT, including an ezetimibe use in agreement with results in similar studies conducted in the time period considered [16], making them eligible for administration of PCSK9i according to Spanish TPR [22].

The main objective for the management of patients with ASCVD is to prevent recurrent CV events. Quick and intensive lowering of LDL-C in these patients is among the most effective therapeutic strategies to achieve this [11,12,13]. We observed that only 11.0% of LLT-treated patients with ASCVD achieved LDL-C goals recommended by the 2019 ESC/EAS guidelines, indicating that more intensive LDL-C-lowering treatment is needed for most patients in this population. Also, we found that 17.9% of Spanish patients with ASCVD receiving optimized oral LLT and at least one LDL-C measurement available in a time period at least 3 months after index date (n = 2600) were eligible for PCSK9i under the current TPR [22] and would benefit from additional lowering of their LDL-C and CV event risk. Reimbursement conditions for PCSK9i in Spain are homogeneous at the national level (TPR). However, additional restrictions on regional and hospital levels often complicate appropriate LLT intensification, especially in patients at very high CV risk that need more aggressive LDL-C lowering [22, 27]. The use of ezetimibe as part of LLT optimization was low in routine clinical practice, even though aligned with similar studies available in the literature and in agreement with the standard practices conducted in the time period considered (2014–2020) [16, 18]. In the few last years there has been a growing trend for using statin and ezetimibe combinations, which could conduct to a better achievement of the LDL-C goals [28].

In European and Spanish clinical practice, adding a PCSK9i such as evolocumab to background oral LLT allowed 2019 LDL-C goals to be achieved in 68% and 72% of patients with very high risk of CV events, respectively, despite very high baseline LDL-C levels in these patient populations [29].

The percentage of uncontrolled patients according to the data analysed in our study was aligned with the data reported by previous studies [16, 18, 30]. These data suggest that high proportions of patients at very high risk of CV events would benefit from more intensive LLT. Additional therapies such as PCSK9i should be introduced to these patients. It is especially noteworthy that even though 17.9% of patients with ASCVD could be eligible for PCSK9i treatment, only 0.9% of patients received them at some point during their treatment.

Use of PCSK9i in the Spanish IQVIA EMR database was low. Some of the reasons for such low use could be that the database contains mainly outpatient records, and the fact that hospital prescriptions were fully registered in the database only from 2018. However, the low use of PCSK9 inhibitors was consistent with the data reported by EUROASPIRE V (0.4%) [18], DA VINCI study (1.2%) [16] and local studies [31], all of them suggesting that there were many patients potentially eligible for PCSK9i under reimbursement conditions that would benefit from more intensive LLT but were not receiving these drugs [22].

The LLT treatment patterns that we observed in our study were consistent with those reported by earlier research [16, 18]. Close to a half of patients with ASCVD receiving LLT (45.4%) were treated with high-intensity LLT (mainly with high-intensity statins). Proportions of patients with MI and UA on high-intensity statins reached 61.9% and 55.5%, respectively. Those percentages were higher than those reported by EUROASPIRE V survey (49.9%) [18] for similar patients. High-intensity LLT comprised mainly of atorvastatin 40–80 mg and rosuvastatin 20–40 mg, in similar relative proportions as reported by the EUROASPIRE V survey (86% in our study versus 72% in EUROASPIRE V and 14% versus 20%, for atorvastatin and rosuvastatin respectively) [18]. In addition, we observed that 11.2% of the patients were on combinations of statins with ezetimibe, this percentage was slightly higher than 8% reported by the EUROASPIRE V survey [18].

In terms of LLT intensity, it is established that the intensity of statins is associated with decreased morbidity and mortality after an acute MI [32,33,34,35]. ESC/EAS guidelines recommend the initiation of high-intensity statins early after admission for all patients with acute coronary syndrome without contraindications or a history of intolerance irrespective of the LDL-C value [10, 13, 36]. In case statins are not enough to reach the recommended LDL-C goals, the guidelines further suggest adding ezetimibe and PCSK9i [11,12,13].

LLT intensification seemed to be conducted during the first weeks after hospital discharge, reaching 2019 target LDL-C reduction goals in 11.0% of patients with ASCVD. This low proportion of controlled patients appeared constant during all the treatment period, denoting signs of therapeutic inertia which leaves many patients at excess risk of recurrent CV events. This therapeutic inertia, together with the lack of adherence of most patients to oral LLT, leads to decreasing control of LDL-C levels over time, as was shown in Fig. 4, as well as to increased risk of death [37, 38]. Addition of ezetimibe was sometimes considered, but addition of PCSK9i seemed to be quite rare. In Spain, the reimbursed use of PCSK9i is focused on well-defined patient profiles, with LDL-C values above 100 mg/dL on optimized statin therapy [22]. Nowadays, according to real-world data in Spain, up to 17.9% of patients would meet PCSK9i reimbursed access conditions. Therefore, there is a clear need for further LDL-C reduction to minimize the potential risk of a new CV event. Introducing PCSK9i to eligible patients will help further reduce their LDL-C and their risk of recurrent CV events.

Major Strengths and Limitations of the Study

The EMR Spanish database is a real-world longitudinal database including routine clinical practice data registered through the informatics systems available in the healthcare centres in Spain. The database provides data directly collected in the electronic systems of the Spanish National Health System, with a high sample size. This gives robustness to the analyses and results, being consistent with other similar studies conducted with similar population profile [16, 18].

The analysis also had some limitations. First, the data registry could be subject to omissions, errors, or divergence in recording practices among clinicians (e.g. 51.5% of the population had missing data related to LDL-C measurements, information about smoking habits was also limited). Second, the data source was focused on outpatient management that could lead to over-representation of the primary care registries over hospital records. Primary care management could result in severe events, such as MI sometimes not being registered in the database. It may also have limited information on hospital prescriptions.

Conclusion

We observed that the proportion of Spanish patients with ASCVD who achieved LDL-C treatment goals recommended by the 2019 ESC/EAS guidelines remained very low indicating the need for further intensification of LLT to lower cardiovascular risk in these very high risk patients. Our study showed that an estimated 17.9% of Spanish patients at very high risk of recurrent CV events were eligible for PCSK9i under the TPR conditions (LDL-C levels exceeding 100 mg/dL despite being on optimized oral LLT); however, only less than 1% of the patients were actually treated with a PCSK9i. Patients who are not treated with a PCSK9i despite being eligible and remain at an unnecessarily excess risk of a recurrent CV event. Intensification of LLT in Spanish clinical practice including wider use of PCSK9i in eligible patients will help reduce cardiovascular risk and prevent recurrent CV events in patients with ASCVD.