Author + information
- Received August 24, 2007
- Revision received November 1, 2007
- Accepted November 14, 2007
- Published online April 15, 2008.
- Bertram Pitt, MD, FACC⁎,⁎ (, )
- Joseph Loscalzo, MD, PhD, FACC†,
- Joseph Yčas, PhD‡ and
- Joel S. Raichlen, MD, FACC‡
- ↵⁎Reprint requests and correspondence:
Dr. Bertram Pitt, 1500 East Medical Center Drive, 3910 Taubman Center, Ann Arbor, Michigan 48109-0366.
Objectives This analysis from the LUNAR (Limiting UNdertreatment of lipids in ACS with Rosuvastatin) study assessed lipid changes 1 to 4 days after onset of acute coronary syndromes (ACS), before initiation of study treatment.
Background Early studies indicated that cholesterol levels decrease significantly after ACS. However, most studies were small or did not measure low-density lipoprotein cholesterol (LDL-C) directly, and many used nonfasting or retrospective data. More recent studies suggest less pronounced changes in cholesterol levels after ACS.
Methods The LUNAR trial is a prospective, multicenter, randomized, open-label study in adults hospitalized for acute ST-segment elevation myocardial infarction (STEMI), non-STEMI, or unstable angina (UA). Blood samples were taken at median times after onset of ACS symptoms of 26 h (Day 1, fasting or nonfasting sample), 43 h (Day 2, fasting sample), and 84 h (Day 4, fasting sample) for direct measurement of serum lipid levels before study treatments were started.
Results Of 507 patients available for analysis, 212 were admitted for STEMI, 176 for non-STEMI, and 119 for UA. The LDL-C levels decreased in the 24 h after admission (from 136.2 to 133.5 mg/dl), followed by an increase over the subsequent 2 days (to 141.8 mg/dl). These changes did not seem to be clinically meaningful. Similar changes were observed for total cholesterol and smaller changes for high-density lipoprotein cholesterol; fasting triglyceride levels did not change.
Conclusions Mean lipid levels vary relatively little in the 4 days after an ACS and can be used to guide selection of lipid-lowering medication. (LUNAR Phase IIIb Study Comparing Rosuvastatin and Atorvastatin in Subjects With Acute Coronary Syndromes; NCT00214630)
Increasing evidence suggests that statins decrease morbidity and mortality when administered early after admission to patients with an acute coronary syndrome (ACS) (1–4). Although current guidelines recommend measurement of serum lipids after admission for patients with an ACS (5,6), less than one-half of these patients have serum lipids measured within 24 h of admission. This is of importance because in-hospital lipid testing and initiation of statin use in patients with an ACS are strongly associated with their use on discharge (7,8).
Many physicians fail to measure serum lipids early after admission for an ACS in part because of their understanding that such measurements are unreliable because of an early decrease in serum total cholesterol (TC) and low-density lipoprotein cholesterol (LDL-C) after ACS. As early as 1971, Fyfe et al. (9) noted apparent decreases in TC during hospitalization of patients with an ACS. Since then, a wide range of alterations in serum lipids has been reported associated with hospitalization for an ACS. For example, Sniderman and Teng (10) found that LDL-C decreased by 18% to 53% in 6 patients with myocardial infarction (MI) between Day 1 and Days 7 to 9, but LDL-C increased by 13% to 32% in 3 other patients. Rosenson (11) cited a general consensus among many studies supporting a large decline in lipid and lipoprotein levels and postulated multiple causes that included an increase in serum triglycerides (TG), changes associated with an acute-phase reaction, and changes in the fasting state. According to this view, reliable measurement of serum lipids might not be achieved until 1 to 2 months after discharge. These earlier data, however, were derived from relatively small populations, and in most circumstances, LDL-C was indirectly determined by calculation based on lipid levels measured from samples taken in varying states of fasting and hydration. Baseline lipid values were often imputed from late post-discharge samples.
Accurate knowledge of baseline lipid levels may affect the initiation of lipid-lowering therapy, the selection of a specific statin and its dosage, and recognition of the potential need for adjunctive lipid therapy, and may influence the patient's willingness to adhere to a recommendation for long-term lipid-lowering therapy.
The LUNAR (Limiting UNdertreatment of lipids in ACS with Rosuvastatin) trial is a prospective, multicenter study in which patients hospitalized for an ACS without statin treatment have a measurement of serum lipids on Days 1, 2, and 4 after the onset of the event, with both direct and indirect measurement of LDL-C. Thus, the LUNAR study provides an opportunity to prospectively examine the changes in serum lipid parameters in a relatively large trial under contemporary conditions.
Patients 18 to 75 years of age who were not on lipid-lowering therapy for at least the preceding 4 weeks were eligible for screening on admission to the hospital within 48 h of onset of ischemic symptoms. Patients could participate if they had unstable angina (UA) or an acute non–ST-segment elevation myocardial infarction (non-STEMI); those with an acute STEMI were eligible if they had undergone successful reperfusion with a thrombolytic agent or primary catheter-based intervention within 12 h of symptom onset. Female patients could not be pregnant and were required to be sterile or using effective contraception to participate in the study.
Patients were excluded for any of the following reasons: uncontrolled diabetes or hypothyroidism; acute liver disease or hepatic dysfunction; elevation of serum creatine kinase not due to myocardial injury; serum creatinine >2.0 mg/dl; pulmonary edema; moderate or severe congestive heart failure; ventricular fibrillation or sustained ventricular tachycardia; prior Q-wave infarct, stroke, sepsis, acute pericarditis, systemic or pulmonary embolus, severe anemia (hematocrit <28%), or participation in another investigational drug study within the previous 4 weeks; coronary artery bypass graft surgery within the past 3 months; or previous percutaneous coronary intervention within the past 6 months.
Each site obtained approval of the study protocol from an institutional review board or an independent ethics committee; all patients provided signed informed consent. The study was conducted in accordance with the ethical principles of the Declaration of Helsinki.
The LUNAR study design
The LUNAR study is a prospective, multicenter, randomized, open-label, 3-arm, parallel-group, 12-week study begun in September 2003 (AstraZeneca study D3560L00021 [4522US/0001]). The primary objective of the LUNAR study is to compare the efficacy of rosuvastatin 20 and 40 mg with that of atorvastatin 80 mg in lowering levels of LDL-C over 6 to 12 weeks of once-daily therapy. Study investigators are blinded to the results of end-point assessments.
Pretreatment lipid assessments
Blood samples were drawn on admission to the hospital (Day 1, fasting or nonfasting sample), after fasting at least 12 h overnight (Day 2), and before randomization to study treatment (Day 4, fasting sample). The date and time of onset of ACS symptoms and of the first blood sample were recorded. Serum levels of LDL-C were measured with the Direct LDL-C Plus process (Roche Diagnostics, Indianapolis, Indiana) at a central facility (Medical Research Laboratories, Highland Heights, Kentucky). Serum levels of high-density lipoprotein cholesterol (HDL-C), TC, and TG were also measured. The LDL-C was calculated indirectly by the Friedewald equation (12) for samples obtained under fasting conditions and for which TG levels were <400 mg/dl. Values for LDL-C obtained by direct measurement correlated well with those calculated indirectly by the Friedewald equation (r = 0.97, slope = 1.0; data from Day 2), with calculated LDL-C levels averaging 11.3 mg/dl lower than directly measured LDL-C levels. Only patients with measurements at all 3 time points were included in this analysis.
Patients must not have used lipid-lowering medication within the previous 4 weeks. Lipid-regulating drugs and enzyme inducers, such as phenytoin, phenobarbital, and carbamazepine, were prohibited to prevent confounding effects on study assessments. Medications were disallowed if they increased the risk of myopathy or caused other safety concerns when coadministered with study medications.
Baseline values for LDL-C, HDL-C, and TC were obtained on Day 1 (fasting or nonfasting). For TG, only fasting measurements were included in the analysis. Data were summarized as means with standard deviations. Analysis of variance was used to examine within-patient differences between various time points. To assess pairwise differences between time point observations, the Scheffé adjustment for multiple comparisons was applied. Statistical significance was defined as p < 0.05. Changes in lipid values were defined as the signed difference between the later and earlier time points, with a negative value indicating a decrease.
Of the 828 patients randomized in the LUNAR trial, 507 patients had direct LDL-C measurements at all 3 time points needed for the present analysis. Demographic characteristics of the analysis population are summarized in Table 1. Blood sampling was performed at median times after onset of ACS symptoms of 26 h (Day 1), 43 h (Day 2), and 84 h (Day 4).
Mean LDL-C, HDL-C, and TC levels are shown in Figure 1. These lipid levels seemed to decrease slightly from Day 1 to Day 2 and to increase again by Day 4 (Table 2). These changes were observed both in the overall population and in each ACS subgroup. Some of the changes in mean LDL-C levels over time were statistically significant, but the magnitudes of the mean changes were small (Table 2). The LDL-C decreased by only 1.70% on the day after admission and increased by Day 4 by 5.01%. The median change between Day 1 and 2 was −2 mg/dl (interquartile range [IQR] −11 to +6 mg/dl), between Days 2 and 4 was 8 mg/dl (IQR −2 to +17 mg/dl), and between Days 1 and 4 was 6 mg/dl (IQR −7 to +18 mg/dl). Similar patterns were found for STEMI and non-STEMI, whereas patients with UA did not have significant changes in the first 24 h after admission (0.81%). Similar changes were observed for total cholesterol, and smaller changes were observed for HDL-C.
Changes in TG levels were analyzed only in patients who gave a baseline blood sample in the fasting state on hospital admission. For this group (n = 249), mean TG levels did not significantly change from Day 1 (183.4 mg/dl) to Day 4 (177.3 mg/dl) (Fig. 2). Although the mean percent change from Day 1 to Day 4 for the overall group was +9.8%, the median percent change was only −0.62%.
Effect of admission delay on LDL-C
Cohorts were defined by the recorded length of time between the onset of symptoms and the time of first blood sample on admission (Day 1). Time intervals were defined as 0 to 12 h, 12 to 24 h, 24 to 36 h, and 36 to 48 h. These cohorts contained 69 patients, 164 patients, 146 patients, and 96 patients, respectively (Fig. 3). Corresponding mean times (symptom onset to first blood sample) for the cohorts were 7.4 h, 18.9 h, 29.3 h, and 42.6 h. Mean LDL-C values in admission samples were similar across cohorts, with slightly higher values in later cohorts. In the cohorts within 1.5 days of symptom onset (0- to 12-h cohort, 12- to 24-h cohort, and 24- to 36-h cohort), there was a small dip in LDL-C from admission (Day 1) to the next in-hospital day (Day 2, range −3.1 to −6.0 mg/dl), followed by an increase on Day 4. Thus, the dip between Day 1 and Day 2 seemed to be associated with the acute hospital admission after the ACS. In the 36- to 48-h cohort, the first 2 measurements were essentially unchanged (+0.3 mg/dl), followed by a small increase on Day 4. Analysis of these cohorts provided no evidence that there was a decrease in LDL-C related to time from onset of the ACS.
Results of this analysis suggest that directly measured serum LDL-C after admission for an ACS changes slightly in a statistically significant but not a clinically meaningful way (+5.01%) from Day 1 to Day 4 (Table 3). Randomization to statin therapy in this study precluded comparative measurement of serum lipid parameters after discharge from the hospital. Nevertheless, the data show that there is relatively little change in any of the measured parameters in any subgroup (Figs. 2 and 3) during the first 4 days after admission for an ACS. This suggests that LDL-C determinations can be calculated reliably from lipid levels on the first available fasting blood sample rather than waiting for discharge or post-discharge blood samples.
When LDL-C levels were analyzed by cohort on the basis of delay from onset of ACS symptoms to admission, mean admission values showed no decrease across time with cohorts ranging from those admitted shortly after the ACS, with a mean time from symptom onset to blood draw of only 7.4 h, to those admitted as late as 36 to 48 h after symptom onset (Fig. 3). This suggests that the small dip in LDL-C was associated with the day after admission (Day 2 of hospitalization), rather than with the time after onset of ACS symptoms. This observation is similar to the slight decrease in LDL-C levels from admission to Day 3 found in patients hospitalized for a variety of complaints, including infectious, respiratory, and gastrointestinal diseases (13). Therefore, the decrease in LDL-C concentration on Day 2 of hospitalization may reflect causes related to hospitalization, such as altered oral intake or intravenous hydration. In addition, in patients with cardiovascular disease, drugs and procedures related to diagnosing or treating ACS are known to affect one or more serum lipid components (14,15).
Early studies of acute MI reported decreases in TC and LDL-C of up to 50% over the time period observed in this study (11). However, as therapeutic intervention has evolved, the magnitude of acute lipoprotein alterations detected has apparently declined with time. In 2001, the MIRACL (Myocardial Ischemia Reduction with Aggressive Cholesterol Lowering) trial (1) reported that LDL-C levels obtained during hospitalization in the placebo arm were 12% lower than those obtained 16 weeks after discharge from the hospital for ACS. However, only 1 sample was reported during hospitalization in MIRACL, precluding any insight into changes in serum lipids during the early post-ACS phase. Several smaller recent studies reporting from 1999 to 2003 have also shown changes in lipid levels after ACS that were less dramatic than those described in earlier reports (16–19). A large-scale review of patient records of admissions for MI found relatively little decrease of LDL-C between samples taken <24 h (120 mg/dl) and >24 h (116 mg/dl) after admission (7). In addition to reflecting an earlier era of care for ACS, the decrease in lipid levels observed in the earlier studies may also relate to their retrospective nature and reliance on imputed baseline lipid levels, the relatively small sample sizes involved in many of them, and the use of derived LDL-C levels based on nonfasted blood samples, which have inherently higher TG levels (and therefore, lower calculated LDL-C levels). The larger recent LATIN (Lipid Assessment Trial–Italian Network) study of patients admitted within 12 h of symptom onset for MI or UA found a mean 7% (UA) to 10% (MI) decrease from admission to the next day in direct-measured LDL-C that persisted until discharge (20). Measurements at 3 months showed that total cholesterol and LDL-C in MI and UA patients returned to levels similar to admission values, suggesting that admission values were representative of true baseline levels. The differences in post-admission decrease between LUNAR and LATIN may be related to differences in patient populations, hospital practices, and treatment patterns in the 2 studies.
A limitation of these data is that although serum lipid values varied relatively little during the early phase post-ACS, the immediate pre-event values for these patients were not available. It is therefore possible that the values observed in this study may be lower than the pre-admission values because of a number of factors, including use of heparin, beta-adrenergic blocking drugs, and other drugs administered during the early phase of an ACS. There also may be an impact on serum lipids of acute-phase reactants, cytokines, and free fatty acids, all of which are elevated during the early phase of an ACS. Nevertheless, the important observation in the present study is that patients currently admitted to the hospital for an ACS now show relatively little variation in serum lipid values over the days after admission. As a consequence, we believe that serum lipids should be measured early after admission for an ACS so that appropriate doses of lipid-lowering therapy may be selected and administered at that time.
In summary, the changes in lipid levels after an ACS seen in this study are much smaller than those noted in older studies. However, they are in keeping with trends in the more recent literature that show less change after an ACS. Our major finding of a small transient decrease on the day after admission is consistent with changes attributable to hospitalization and to treatments typically initiated with ACS.
In conclusion, reliable knowledge of serum lipid levels early after admission to the hospital for an ACS should facilitate initiation of lipid-lowering therapy (most likely involving a statin), allow a more rational selection of drug dosage, and identify the potential need for adjunctive lipid-altering therapy. The knowledge that mean serum LDL-C and other lipid parameters vary relatively little over the early days after admission for an ACS should provide an impetus to measure lipid parameters early after admission, thus providing a rational basis for clinical decisions about lipid-lowering therapy.
The authors thank Drs. Annemarie Clegg and Michael Theisen, from Scientific Connexions, who provided editorial assistance with figures and references.
The LUNAR study and the current analysis were supported and funded by AstraZeneca LP. Dr Pitt has served as a consultant to AstraZeneca, Novartis, and Pfizer. Drs Yčas and Raichlen are employed by AstraZeneca LP.
A preliminary version of this analysis was presented as a poster at the 55th Annual Scientific Session of the American College of Cardiology, Atlanta, Georgia, March 11 to 14, 2006.
- Abbreviations and Acronyms
- acute coronary syndrome(s)
- high-density lipoprotein cholesterol
- interquartile range
- low-density lipoprotein cholesterol
- ST-segment elevation myocardial infarction
- total cholesterol
- unstable angina
- Received August 24, 2007.
- Revision received November 1, 2007.
- Accepted November 14, 2007.
- American College of Cardiology Foundation
- Schwartz G.G.,
- Olsson A.G.,
- Ezekowitz M.D.,
- et al.,
- Myocardial Ischemia Reduction with Aggressive Cholesterol Lowering (MIRACL) Study Investigators
- Antman E.M.,
- Anbe D.T.,
- Armstrong P.W.,
- et al.,
- American College of Cardiology; American Heart Association; Canadian Cardiovascular Society
- Braunwald E.,
- Antman E.M.,
- Beasley J.W.,
- et al.,
- American College of Cardiology,
- American Heart Association,
- Committee on the Management of Patients With Unstable Angina
- Rosenson R.S.
- Friedewald W.T.,
- Levy R.I.,
- Fredrickson D.S.
- Dupuis J.,
- Tardif J.-C.,
- Cernacek P.,
- Théroux P.