The registry of patients admitted for suspected ACS to the coronary care unit of our center (a tertiary hospital with a referral population of more than 400 000 people) was reviewed between January 1st, 2010, and April 30st, 2017. This systematically and prospectively compiled registry collected their main demographic data, risk factors, comorbidities, previous treatments, clinical presentation, clinical course during admission, results of complementary examinations, including the fasting lipid profile obtained on the morning of the first working day after admission, and treatments during admission and at discharge, according to established recommendations.21 The exclusion criteria were patients transferred from another center, those who had ACS while they were hospitalized for another reason, those whose early lipid profile was unavailable because they were quickly transferred to another hospital or for any other cause, and those whose final diagnosis was not ACS. The study was approved by the clinical research ethics committee of our hospital.

LDL-C concentrations were calculated using the Friedewald formula22 or were directly measured after ultracentrifugation in individuals with significant hypertriglyceridemia. Patients were deemed to have adequate lipid control at admission if their LDL-C concentrations were within the limits established by the applicable European guidelines on cardiovascular prevention10 or dyslipidemia management.11 According to these recommendations, the LDL-C target is <70mg/dL (or a ≥ 50% reduction if the baseline level is 70-135mg/dL) in patients with documented cardiovascular disease and in those in primary prevention considered at very high risk due to the presence of diabetes with target organ damage or of an associated major cardiovascular risk factor, severe chronic kidney disease (estimated glomerular filtration rate <30mL/min/1.73 m2), or an estimated 10-year risk of cardiovascular mortality ≥ 10%; the LDL-C target is <100mg/dL (or a ≥ 50% reduction if the baseline level is 100-200mg/dL) in patients considered at high risk due to the presence of a poorly controlled risk factor, diabetes mellitus without the above criteria, moderate chronic kidney disease (estimated glomerular filtration rate 30-59mL/min/1.73 m2), or an estimated 10-year risk of cardiovascular mortality of 5% to 10%. The target is <115mg/dL in the other patients with an indication for lipid-lowering therapy due to a moderate estimated risk. Finally, there is no defined target in the rest of the population, although an LDL-C ≥ 155mg/dL is considered inappropriate, unless the estimated risk is extremely low. Due to the limitations in the available information, only the absolute LDL-C targets were considered in the present study because the values prior to therapy initiation were not available for most patients undergoing lipid-lowering therapy and no attempt was made to retrospectively estimate the 10-year risk of cardiovascular mortality.

Categorical variables are expressed as numbers and percentages and continuous variables as median [interquartile range]. Two categorical variables were compared using the chi-square test, whereas 2 continuous variables were compared using the Mann-Whitney U test. The presence of temporal trends in lipid target adherence was evaluated by comparing 4 consecutive time periods. Comparisons among these periods and among other ordinal variables, such as cardiovascular risk categories, were made using the chi-square test for linear trend for categorical variables and by the Jonckheere-Terpstra test for continuous variables. Variables independently associated with inadequate lipid control at admission were identified through multivariable logistic regression analysis with step-by-step elimination that included demographic variables, other cardiovascular risk factors, and the main comorbidities. Model calibration was assessed using the Hosmer-Lemeshow test and its discriminative capacity using the C statistic. A sensitivity analysis was performed by excluding patients who were admitted on Friday or Saturday and therefore lacked a lipid profile in the first 24hours of admission. However, these patients were kept in the study after verification that the results were practically unchanged. Through multivariable logistic regression analysis, the association between lipid control and hospital mortality was analyzed and the population attributable risk was calculated according to the methodology described.23 All analyses were performed with SPSS statistical software. P values <.05 were considered significant.

Most studies evaluating the degree of lipid control have focused on stable patients. In recent surveys conducted in Europe or in other regions around the world, the percentage of patients with stable coronary heart disease and LDL-C concentrations <70mg/dL ranged between 7% and 30%.15,16,19,20 In studies restricted to patients being treated with statins, with or without documented cardiovascular disease, optimal control percentages ranged from 27% to 52%.13,14,17 In Spain, analysis of the Spanish cohort of one of these last studies obtained results that were in line with the general figures.24 In addition, also in Spain, LDL-C percentages within the target range were recently reported in 26% of 1103 patients with stable coronary heart disease25 and in 41.5% of 392 young people with familial hypercholesterolemia.18

Although inadequate lipid control has been associated with a higher incidence of ACS,5,6,8,26 dyslipidemia has often been considered as a dichotomous variable and few studies have analyzed the degree of baseline lipid control in patients admitted for ACS. In a large North American registry that included patients admitted for stable or unstable coronary disease between 2000 and 2006, 17.6% of the 136 905 patients whose LDL-C was measured (59% of the total) had levels <70mg/dL.12 Recently, the DYSIS-II registry27 included, in addition to patients with stable coronary disease, 3867 patients admitted for ACS in 2012 and 2013 in hospitals from 18 countries in Asia, Europe, and the Middle East; 18.9% had LDL-C concentrations at admission <70mg/dL. Finally, in a Greek study of 687 patients with ACS, patients with LDL-C concentrations in line with the recommended targets comprised less than 20% of the highest risk groups and between 30% and 40% of the lowest risk groups.28

This study provides contemporary information on the degree of LDL-C control in unselected patients admitted for ACS in Spain. No data had been available on this topic because Spanish hospitals were not involved in the DYSIS-II study. Compared with the studies mentioned, the proportion of patients with LDL-C levels within the recommended target range was higher in the present study, but still insufficient. Some data provide hope that the situation will improve. First, and in line with other registries,12,29,30 there was an increase, although small, in the proportion of patients with optimal lipid control in recent years. Second, the percentage of patients with extreme deviations from the target was relatively low. Finally, the lipid-lowering therapy was far from optimal in a considerable number of patients. As already observed,5,7,13,19,27 the percentage of patients with adequate LDL-C concentrations should be significantly increased by a more aggressive drug therapy based on the prescription of appropriate doses of potent statins to patients with high or very high risk, close analytical monitoring, and the use of combination therapies when monotherapy is insufficient. Lipid control can be improved with simple interventions,31,32 which must be maintained over time to avoid loss of treatment adherence among patients.33

The degree of LDL-C control was inversely associated with the patients’ cardiovascular risk. This association, consistent with previous observations,14,17,28 is partly because the targets are more stringent for patients with higher risk, although it may also be more difficult for these patients to achieve strong and persistent LDL-C reductions due to worse adherence to treatments and healthy lifestyles or for other reasons. Specifically, poor LDL-C control was independently associated with dyslipidemia, smoking, diabetes, and excess weight, whereas statin therapy was associated with better control. These results also partly agree with previous observations,12–14,18,27 although diabetes has been associated with better lipid control in some studies.12,25,27 The results identify a subgroup of patients (with excess weight or these other risk factors) that warrant priority attention.

Although the patients were enrolled from a single center, their origin was diverse, not only because they belonged to the large catchment area of our hospital, but also because many patients admitted in line with the infarction code program came from remote regions. On the other hand, the series was large and includes almost as many patients with ACS as the largest multinational study published to date.27 Thus, the results are probably representative of the population of our country.

The lipid control targets were based on cardiovascular risk stratification of the patients according to the applicable European guidelines.10,11 Although, the treatment guidelines for dyslipidemia of 2011 were in force during most of the analyzed period,9 these recommendations were ambitious because they considered all patients with type 2 diabetes or moderate renal dysfunction as being at very high risk (with an LDL-C target <70mg/dL). Given that these targets were substantially reduced in subsequent guidelines and that there is a delay between recommendation publication and implementation, we believe that the current targets provide a more realistic and up-to-date view of patients’ lipid control.

As mentioned, we decided not to exclude patients admitted on Friday and Saturday—most of whom did not undergo an analysis within the first 24hours of admission—after verifying that their exclusion did not modify the results. This observation contrasts with some previous findings34 but is in line with those of the largest relevant study, which showed that LDL-C values do not vary significantly during ACS admission.35 In our study, low LDL-C values at admission were associated with worse clinical presentation and higher hospital mortality. This association may have influenced the results and indicates that, in the most severe patients, the LDL-C values at admission are probably not representative of the degree of prior control. Finally, a limitation in the available data prevented calculation of the 10-year cardiovascular risk and analysis of the association of the degree of lipid control with potential variables of interest, such as the patients’ psychological profile, therapeutic adherence, and statin potency or dose.