1 INTRODUCTION

Cardiovascular disease (CVD) is a leading cause of mortality worldwide.1 Treatments that reduce low-density lipoprotein cholesterol (LDL-C) levels represent an important strategy to reduce cardiovascular (CV) events, as well as mortality.2 Moreover, aggressive therapy to decrease LDL-C levels provides an even greater benefit for patients with established atherosclerotic cardiovascular disease (ASCVD),3-7 particularly for patients with cardiovascular heart disease (CHD). The effects of sustained LDL-C reduction on CV events from other territories have received little attention. Over time, successive evidence has led to the recommendation of more stringent LDL-C target levels for ASCVD patients, decreasing from 100 mg/dl to the current recommendation by the 2019 European Society Cardiology (ESC) and European Atherosclerosis Society (EAS) Guidelines of levels <55 mg/dl for ASCVD patients or levels <40 mg/dl for those who experience a recurrent vascular event within 2 years.8 However, scientific evidence is often poorly implemented when it is incorporated into clinical practice and surveys from multiple countries indicated that targeted cardiovascular risk management practices remain far from optimal.9 Despite guideline recommendations and availability of more intensified lipid-lowering therapy, many ASCVD patients do not achieve LDL-C goals.10-12

Randomized clinical trials are considered the “gold standard” for assessing the effects of decreasing LDL-C levels among patients with ASCVD. However, these studies had restrictive inclusion criteria, and are conducted in a highly controlled setting that cannot be replicated in clinical practice. In this way, reports of clinical efficacy in a real-world setting with the full spectrum of ASCVD are scarce, particularly for patients with peripheral vascular disease.12-15 However, these studies analysed large populations and have some limitations. These reports provide evidence from analysis of large populations, but also have some limitations. Most observational studies are retrospective studies collected from electronic national databases analysing a short number of factors, excluding other associated with the development of cardiovascular events. Most often studies provide cross-sectional information on LDL-C with no information on long-term sustained LDL-C levels. Additionally, patients with peripheral artery disease (PAD) and a high risk of recurrence are often underrepresented or simply absent.16 As such, data on the actual translation of guideline recommendations for ASCVD patients and derived real-life outcomes remain uncertain.

The Factores de Riesgo y ENfermedad Arterial (FRENA) Registry was a study that prospectively gathered data on the clinical management and outcome of patients with atherosclerotic vascular diseases in several Spanish hospitals. FRENA was a multicentre, observational registry of consecutive stable outpatients with symptomatic ischaemic arterial disease of the heart, brain and/or major peripheral arteries. Data from this registry have been shown as valuable tools for studying outcomes of in a diverse patient population such as influence of body weight, smoking habits, alcohol consumption, glucose control, renal function, proton pump inhibitors, systolic blood pressure, anticardiolipin antibodies or lipoprotein (a) on clinical outcomes.17-25 The aim of the current study was to assess the effect of sustained LDL-C levels <70 mg/dl on the incident rate of subsequent ischaemic arterial events among a cohort of ASCVD followed-up for 5 years.

2 PATIENTS AND METHODS

The FRENA registry was a Spanish prospective study from March 2003 to October 2018. Participating hospitals in the FRENA registry prospectively enrolled consecutive stable outpatients with symptomatic artery disease with an indexing event in the previous 3 months. The original study established a minimum requirement of 1-year follow-up. For the purpose of this analysis, only subjects with a complete follow-up of 5 years, or a recurrent ischaemic event or death (whichever occurred first), were included. Symptomatic artery disease was defined as an acute coronary episode (unstable angina requiring revascularization or acute myocardial infarction); ischaemic cerebrovascular (transient ischaemic attack or ischaemic stroke) or peripheral artery disease (intermittent claudication with an ankle-brachial index <0.9, previous limb revascularization or amputation). Patients were excluded if they did not attend to 3 or more medical appointments, were lost to follow-up or if they were enrolled in therapeutic clinical trial. All patients provided informed consent prior to their participation in the registry, according to the requirements of the ethics committee within each hospital. Reporting of the study conforms to broad EQUATOR guidelines.26

2.1 Study design

Treatment goals were defined according to recommendation of contemporary international Guidelines. Consequently, patients were categorized into two groups according to LDL-C levels: LDL-C <70 mg/dl or ≥70 mg/dl. The primary outcome was the incidence of subsequent major adverse cardiovascular events (MACE) during follow-up. MACE were defined as the composite of major coronary events (myocardial infarction or urgent coronary revascularization), ischaemic stroke (stroke with neurologic sequelae or urgent revascularization), amputation (leg or foot/toes amputations) or related death (whichever occurred first). Transient ischaemic attacks or stable angina were not considered as outcomes.

2.2 Definitions and laboratory tests

Quantification of blood tests was performed by using local laboratory standard methods in each participant hospital. LDL-C level was estimated by the Martin–Hopkin formula.27, 28 The average of LDL-C determinations of each medical visit was considered before the occurrence of a new MACE. Glomerular filtration rate was estimated with the Cockcroft–Gault formula. All participant hospitals carried out identical quality controls according to guidelines of Spanish Ministry of Health.

Diabetes was considered for patients with fasting plasma glucose >126 mg/dl, a previous diagnosis of diabetes or treatment with insulin or oral antidiabetic agents. Hypertension was defined as blood pressure above 140/90 mmHg, previous diagnosis of hypertension or treatment with antihypertensive medications.

Lipid-lowering therapy was defined as: (a) high-intensity statin: patients receiving 20–40 mg rosuvastatin or 40–80 mg atorvastatin; and (b) moderate-intensity statin: patients receiving lower doses of rosuvastatin (<20 mg) or atorvastatin (<40 mg) or any other statin or other lipid-lowering drugs included ezetimibe or fibrates. No patients were receiving proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitor therapy, according to the date of the inception of the cohort.

Cardiovascular event-free survival was calculated for all patients. Patients were censored after the occurrence of the first arterial ischaemic event (MACE), death or last follow-up visit, whichever occurred first. Odyssey MACE outcome criteria were considered,29 with the addition of ischaemic lower limb amputation. (1) Coronary: Myocardial infarction was defined as the presence of typical chest pain in combination with a transient increase in creatine kinase-MB or troponin and/or typical electrocardiogram signs (development of pathologic Q-waves or ST-segment elevation or depression); (2) Cerebrovascular: Ischaemic stroke was diagnosed if the patient had a focal neurological deficit lasting >24 h, and had an acute ischaemic nonhaemorrhagic cerebrovascular lesion on brain computed tomography or magnetic resonance imaging; (3) Peripheral: Lower limb amputation if secondary to chronic or acute ischaemic lesions.

2.3 Follow-up

A detailed history was performed by a hospital physician investigator in each medical appointment. Co-morbid conditions were recorded, including a previous history of coronary, cerebrovascular or peripheral artery disease, diabetes, hypertension, chronic lung disease, heart failure, cancer as well as smoking status and alcohol consumption. Physical examination included weight, height, BMI, heart rate and blood pressure levels on standard conditions, after 5 min of rest. An electrocardiogram was also recorded. After the initial visit, patients were followed-up at 4-month intervals in the outpatient clinic during first 2 years and then every 6 months until the end of a 5-year follow-up period. At all visits, any change in medical history and data from physical examination were recorded, with special attention to lifestyle habits; blood pressure measurement; laboratory tests including lipid basic evaluation; drug therapy and clinical outcomes. Attending clinicians were allowed to prescribe all appropriate medications according to national guidelines and clinical judgement. All patients included were followed for up to 5 years in the out-patient hospital clinics (or until the appearance of new MACE or death if they occurred earlier). Most outcomes were classified as reported by the clinical centres. A central adjudicating committee revised all cases with uncertain classification (<10% of events).

2.4 Data collection

Data were recorded on a computer-based case report form at each participating hospital and submitted to a centralized coordinating centre through a secure website. Patient identities remained confidential as they were identified by a unique code assigned by the study coordinating centre, which was responsible for all data management. Data quality was regularly monitored and documented electronically to detect inconsistencies or errors, which were resolved by the local coordinators. Data quality was also monitored by periodic visits to participating hospitals, by a contract research organization, which compared the medical records with the data in the web. A data audit was performed at 6-month intervals.

2.5 Statistical analysis

Categorical variables were compared using the Chi-square test (two-sided) and Fisher's exact test (two-sided). Hazard ratios (HR) and corresponding 95% confidence intervals (CI) were calculated, and p value < 0.05 was statistically significant. Incidence rates were calculated as cumulative incidence (events/100 patient-years) and compared using the rate ratio. Recurrent MACE cumulative incidence according to LDL-C levels is shown as an adjusted multivariable Cox regression plot. The association of independent variables with the occurrence of new ischaemic events was evaluated by univariate Cox models. All variables achieving a significance level of 0.1 on univariate analysis were considered for inclusion in the multivariate Cox model for predictors of subsequent ischaemic events and mortality. Variables in the model were as follows: older age (>65 years), sex, territory of the first CV event, early familiar CV disease, systolic blood pressure, comorbidity (heart failure, atrial fibrillation, cancer and chronic bronchitis), LDL-C levels, triglycerides, creatinine clearance and pharmacologic therapy (statins, other lipid-lowering therapy, proton pump inhibitors, diuretic, beta-blockers and anticoagulants). Causes of death according to LDL-C levels were also compared by the Chi-square test or Fisher's exact test. Statistical analyses were conducted with IBM® statistical program spss® Statistics V.24 (IBM Corporation).

3 RESULTS

FRENA registry recruited 5321 patients, 705 of whom were lost or did not complete requirements in follow-up during first year. Of the remaining 4616 cases, 5-year follow-up was attained for 1657 participants, of whom 210 did not attend to 3 or more medical appointments, 187 did not have a full plasma lipid profile and 78 were excluded for missing critical data during follow-up. Finally, 1182 patients were included in analysis for a total follow-up of 5000 patient years (Figure 1). The mean number of LDL-C measurements over follow-up was 13.4 ± 2.3 for each participant. Among them, 172 (14.5%) had LDL-C <70 mg/dl (mean ± standard deviation (SD), 59 ± 10 mg/dl) and 1010 (85.5%) had LDL-C >70 mg/dl (mean ± SD, 109 ± 25 mg/dl), a difference of 50 mg/dl compared with patients in the LDL-C <70 mg/dl. Mean age and sex were similar for patients with LCL-C <70 mg/dl and >70 mg/dl (Table 1). Triglycerides were lower for patients with LDL-C <70 mg/dl compared to LDL-C ≥70 mg/dl: 102 ± 46 and 134 ± 70, respectively, p < 0.001. No other differences were present between both groups with respect to vascular risk factor, underlying diseases, clinical presentation, physical examination or therapies including high-intensity statin and other lipid-lowering drugs.

Participant selection process

TABLE 1. Baseline characteristics of the at-entry population and mean analytical levels in follow-up LDL-C < 70 mg/dl LDL-C > 70 mg/dl p-value Patients, N 172 1010 Clinical characteristics Mean age (years ± SD) 68 ± 13 68 ± 12 0.55 Gender (male) 123 (71%) 729 (72%) 0.85 Body mass index (Kg/m2 ± SD) 28 ± 5 28 ± 4 0.99 Vascular risk factor Hypertension 116 (67%) 681 (67%) 1.00 Diabetes 55 (32%) 396 (39%) 0.07 Current smokers 29 (17%) 169 (17%) 0.83 History of family early vascular events 25 (14%) 131 (13%) 0.63 First ischaemic event 85 (49%) 495 (49%) 0.93 Underlying diseases Cancer 11 (6.4%) 77 (7.6%) 0.64 Chronic lung disease 29 (17%) 168 (17%) 0.91 Heart failure 17 (10%) 100 (10%) 1.00 Atrial fibrillation 25 (15%) 96 (10%) 0.09 Any of the above 61 (36%) 339 (34%) 0.66 Clinical presentation Coronary artery disease 67 (39%) 374 (37%) 0.67 Cerebrovascular disease 46 (27%) 314 (31%) 0.28 Peripheral artery disease 62 (36%) 359 (35%) 0.93 Physical examination, in follow-up Mean SBP levels (mm Hg) 137 ± 23 138 ± 23 0.57 Mean DBP levels (mm Hg) 75 ± 13 75 ± 13 0.92 Mean analytical levels, in follow-up Total cholesterol (mg/100 ml) 128 ± 19 181 ± 30 <0.001 LDL cholesterol (mg/100 ml) 59 ± 10 109 ± 25 <0.001 HDL cholesterol (mg/100 ml) 50 ± 14 48 ± 13 0.27 Triglycerides (mg/100 ml) 102 ± 46 134 ± 70 <0.001 HbA1c (%)a 5.9 ± 1.7 5.8 ± 1.6 0.06 Creatinine Clearance (ml/min) 74 ± 31 73 ± 31 0.67 Drugs Diuretics 65 (38%) 402 (40%) 0.67 Beta blockers 73 (42%) 399 (39%) 0.50 ACE or ARA antagonists 116 (67%) 690 (68%) 0.86 Calcium antagonists 44 (26%) 287 (28%) 0.46 Antiplatelets 150 (87%) 898 (89%) 0.52 Clopidogrel 72 (42%) 401 (40%) 0.61 Anticoagulants 32 (19%) 182 (18%) 0.83 Statins 156 (%) 884 (%) 0.26 High-intensity statins 135 (86%) 733 (83%) 0.29 Moderate-intensity statins 21 (6.5%) 151 (17%) Other lipid lowering 70 (41%) 353 (35%) 0.09 Insulin 25 (14%) 142 (14%) 0.91 Oral antidiabetics 45 (26%) 268 (26%) 1.00 Abbreviations: ACE, angiotensin converse enzyme; ARA, aldosterone receptor antagonists; BMI, body mass index; DBP, diastolic blood pressure; HbA1c, glycated haemoglobin; HDL, high-density lipoprotein; LDL, low-density lipoprotein; SBP, systolic blood pressure; SD, standard deviation.

After 5 years of follow-up, MACE occurred in 252 patients (21%); 5.27 events per 100 person-years (95% CI 4.64–5.95) and 206 (17%) died (Table 2). Patients with LDL-C <70 mg/dl exhibited a lower rate of MACE (rate ratio [RR]: 0.61; 95% CI: 0.39–0.92). MACE reduction was driven by patients with initial coronary heart disease (RR: 0.51; 95% CI 0.25–0.94) and initial limb amputation (RR: 0.35; 95% CI: 0.10–0.88), with no differences for initial ischaemic stroke. Death rate did not differ between achieved LDL-C groups. In our cohort, the overall mortality was 4.30 (95% CI, 3.74–4.92) per 100 person-years. After adjustment for age, gender, prior cardiovascular event, history of early cardiovascular disease, hypertension, heart failure, atrial fibrillation, cancer, chronic lung disease, triglycerides, creatinine clearance and drug therapy, the HR for subsequent MACE outcome (HR 0.61, 95% CI 0.39–0.93) was significantly lower for patients with LDL-C <70 mg/dl versus ≥70 mg/dl, whereas no difference was observed across LDL- C categories for all-cause death (Table 3). By univariate analysis, several factors were associate with mortality, age, glycated haemoglobin (HbA1c) >7%, initial CHD, heart failure, cancer, triglycerides >200 mg/dl, creatinine clearance <60 ml/min and anticoagulant therapy. The most frequent causes of death are shown in Table 4. There were no significant differences between subgroups according to LDL-C levels for any of them. Estimates of efficacy over time that were based on adjusted multivariable Cox regression plot (Figure 2) show the probability of recurrent MACE in two groups according to the mean levels of LDL-C (HR was significantly lower for patients with LDL-C <70 mg/dl [0.61 (0.39–0.93)]). Patients with LDL-C level <70 mg/dl occurred a lower number of vascular recurrences, 23 (13.4%) compared to those with LDL-C higher 70 mg/dl, 229 (22.7%). Cardiovascular deaths occurred in 10 (5.8%) and 72 (7.1%) and all-cause death was 33 (19.0%) and 173 (17.1%) respectively.

TABLE 2. Incidence of subsequent ischaemic events or death according to mean LDL-cholesterol levels in 5 years follow-up LDL-C < 70 mg/dl LDL-C ≥ 70 mg/dl Rate ratio (95% CI) p-value N N per 100 patient-years (95% CI) N N per 100 patient-years (95% CI) All patients, n = 1182 172 1010 Follow-up (years) 671 4111 MACE 23 3.42 (2.17–5.14) 229 5.57 (4.87–6.34) 0.61 (0.39–0.92) 0.019 Overall death 33 4.91 (3.38–6.90) 173 4.20 (3.60–4.88) 1.16 (0.79–1.67) 0.409 CHD, n = 441 67 374 Follow-up (years) 278 1520 MACE 10 3.59 (1.72–6.61) 107 70.39 (5.76–8.50) 0.51 (0.25–0.94) 0.029 Overall death 11 3.95 (1.97–7.08) 47 3.09 (2.72–4.11) 1.28 (0.63–2.40) 0.456 CVD, n = 354 48 306 Follow-up (years) 152 1260 MACE 9 5.92 (2.70–11.24) 59 4.68 (3.56–6.04) 1.26 (0.59–2.46) 0.500 Overall death 9 5.92 (2.70–11.24) 52 4.12 (3.08–5.41) 1.43 (0.66–2.81) 0.322 PAD, n = 387 59 328 Follow-up (years) 241 1331 MACE 4 1.66 (0.44–4.24) 63 4.73 (3.63–6.05) 0.35 (0.10–0.88) 0.022 Overall death 13 5.39 (2.87–9.22) 74 4.73 (3.63–6.05) 1.14 (0.60–2.02) 0.650 Abbreviations: ACS, acute coronary syndrome; CAD, coronary artery disease; CVD, cerebrovascular disease; LDL-C, low-density lipoprotein cholesterol; MACE, major adverse cardiovascular events; PAD, peripheral artery disease. TABLE 3. Predictors for subsequent ischaemic events and mortality MACE All-cause death Clinical characteristics Age >65 years 1.16 (0.85–1.18) 2.57 (1.68–3.93)‡ Gender (males) 1.21 (0.89–1.62) 1.26 (0.90–1.77) First CV event 1.33 (0.96–1.86) Vascular risk factor Early familiar CV disease 1.73 (0.98–3.02) Diabetes 1.35 (0.99–1.84) A1cHb>7% 1.63 (1.09–2.44)* Hypertension 1.06 (0.75–1.50) SBP >140 mmHg 1.07 (0.81–1.42) 0.79 (0.59–1.05) Clinical presentation Coronary artery disease 0.96 (0.70–1.30) 0.68 (0.47–0.98)* Peripheral artery disease 1.15 (0.81–1.62) 1.30 (0.87–1.95) Underlying diseases Heart failure 2.11 (1.47–3.04)‡ 2.18 (1.50–3.18)‡ Cancer 1.67 (1.11–2.51)* Chronic lung disease 1.21 (0.85–1.73) Mean laboratory levels in follow-up LDL-cholesterol <70 mg/dl 0.61 (0.39–0.93)* 0.70 (0.48–1.02) Triglycerides >200 mg/dl 1.59 (1.12–2.27)* Creatinine clearance <60 ml/min 1.17 (0.87–1.58) 1.43 (1.06–1.93)* Drugs Diuretic 1.11 (0.83–1.47) 0.92 (0.69–1.24) Anticoagulants 0.49 (0.32–0.75)† Insulin 1.01 (0.69–1.50) High-intensity statin 0.08 (0.03–0.23)‡ 0.70 (0.48–1.02) Abbreviations: CV, cardiovascular; HbA1c, glycated haemoglobin; LDL-C, low-density lipoprotein cholesterol; SBP, systolic blood pressure. Comparisons: *p < 0.05; †p < 0.01; ‡p < 0.001. TABLE 4. Causes of death according to LDL-C levels LDL < 70 mg/dl LDL ≥ 70 mg/dl Patients, N 172 1010 All-cause death 33 (19%) 173 (17%) Cardiovascular death 10 (5.8%) 72 (7.1%) Heart failure 4 (2.3%) 29 (2.9%) Sudden, unexpected 3 (1.7%) 23 (2.2%) Arrhythmia 0 5 (0.5%) Cerebral oedema 1 (0.5%) 13 (1.3%) Mesenteric ischaemia 1 (0.5%) 0 Abdominal aneurysm 1 (0.5%) 2 (0.2%) Noncardiovascular death 23 (13.4%) 101 (10.0%) Malignancy 7 (4.0%) 22 (2.2%) Infection 3 (1.7%) 13 (1.3%) Bleeding 1 (0.5%) 8 (0.8%) Chronic lung disease 0 6 (0.6%) Unknown 6 (3.5%) 31 (3.1%) Others 6 (3.5%) 21 (2.1%)