Statin therapy moderately increases high-density lipoprotein cholesterol (HDL-C) levels. Contrary to this expectation, a paradoxical decrease in HDL-C levels after statin therapy is seen in some patients. We evaluated 724 patients who newly started treatment with statins after acute myocardial infarction (AMI). These patients were divided into 2 groups according to change in HDL-C levels between baseline and 6 to 9 months after initial AMI (ΔHDL). In total, 620 patients had increased HDL-C levels and 104 patients had decreased HDL-C levels. Both groups achieved follow-up low-density lipoprotein cholesterol levels <100 mg/dl. Adverse cardiovascular events (a composite of all-cause death, myocardial infarction, and stroke) have more frequently occurred in the decreased HDL group compared with the increased HDL group (15.4% vs 7.1%, p = 0.01). Multivariate analysis showed that decreased HDL, onset to balloon time, and multivessel disease were the independent predictors of adverse cardiovascular events (hazard ratio [HR] 1.95, 95% confidence interval [CI] 1.08 to 3.52; HR 1.05, 95% CI 1.01 to 1.09; and HR 2.08, 95% CI 1.22 to 3.56, respectively). In conclusion, a paradoxical decrease in serum HDL-C levels after statin therapy might be an independent predictor of long-term adverse cardiovascular events in patients with AMI.
A low level of high-density lipoprotein cholesterol (HDL-C) is a well known and an important risk factor for cardiovascular disease. Results from several epidemiologic studies have shown a 2% to 3% decrease in future cardiovascular events for every 1 mg/dl increase in HDL-C levels. Thus, a low HDL-C level is a residual risk factor for cardiovascular disease independent of low-density lipoprotein cholesterol (LDL-C) levels even after LDL-C is at the target level. Some studies suggest that increasing HDL-C levels is associated with a reduction in atheroma progression as measured in the coronary and carotid arteries. In addition to lowering LDL-C levels, statin therapy also moderately increases HDL-C levels. Contrary to this expectation, patients sometimes experience a paradoxical decrease in HDL-C levels after statin therapy. The purpose of our study was to analyze the relation between the paradoxical HDL-C decrease after statin therapy and adverse cardiovascular events in patients with acute myocardial infarction (AMI) who underwent PCI.
Methods
The Nagoya Acute Myocardial Infarction Study (NAMIS) was a prospective, multicenter, observational study of AMI with 1,144 patients enrolled from 18 collaborating hospitals in central Japan from January 2004 to December 2012. Precise protocols of NAMIS were described in elsewhere. In brief, NAMIS was designed to assess clinical variables, therapeutic procedures, and subsequent clinical events in patients with AMI. The eligibility criteria were admission to a participating hospital for AMI and successful reperfusion within 24 hours of symptom onset. The independent Endpoint Evaluation Committee, which was blinded to all patient clinical information, strictly evaluated all events.
In this study, we excluded patients with statin therapy on admission, those without statin therapy at discharge, those with therapies that increase HDL-C levels, such as fibrates and pioglitazone, and those without follow-up lipid profiles. As a result, we focused on 817 patients who were newly started on statins after admission. This strategy provided plasma lipid levels for each patient before statin therapy was started and at 6 to 9 months after initial AMI. Change in HDL-C levels was calculated as HDL-C levels at 6 to 9 months follow-up minus baseline HDL-C levels. Similar estimates were also completed for the change in LDL-C and plasma triglyceride (TG) levels. Finally, 724 eligible patients were included in this subgroup analysis. These patients were divided into 2 groups according to change in HDL-C levels (ΔHDL): 620 patients with ΔHDL ≥0 in the increased HDL group and 104 patients with ΔHDL <0 in the decreased HDL group. Written informed consent was obtained from all patients. The study protocol was approved by the ethics committee of each hospital.
In this study, adverse cardiovascular events were defined as a composite of the following complications: all-cause death, recurrent myocardial infarction, and stroke. A diagnosis of stroke was made only when a prolonged neurologic deficit was present with documentation on imaging, such as computed tomography or magnetic resonance imaging. We also collected data on incidence of cardiac death, congestive heart failure, and target vessel revascularization. The left ventricular ejection fraction was calculated by echocardiography using the Teichholz formula at discharge. We obtained creatine kinase levels on admission and every 6 hours until 24 hours after PCI. The estimated glomerular filtration rate (eGFR) based on serum creatinine was calculated using the 3-variable equation, proposed by the Japanese Society of Nephrology (eGFR [ml/min/1.73 m 2 ] = 194 × serum creatinine [−1.094] × age [−0.287] × 0.739 [if female]).
Statistical analysis was performed using the SPSS 17.0 software program (SPSS inc, Chicago, IL). Data were expressed as mean value ± SD or median and interquartile ranges for continuous variables and as percentages for discrete variables. The Student unpaired t test or Mann-Whitney U test was used to compare continuous variables between groups. Categorical variables were compared by chi-square test or Fisher’s exact test, as appropriate. Event-free survival curves for adverse cardiovascular events were constructed by means of the Kaplan-Meier method. Survival probabilities of the 2 groups were compared with the log-rank test. Independent predictors for the occurrence of adverse cardiovascular events were determined by Cox proportional hazard regression using stepwise selection with entry and exit criteria of p <0.05 and p <0.10, respectively. Variables that were evaluated in the multivariate Cox regression analysis included those with significant association in the univariate analysis and also those without statistical significance in the univariate analysis but with prognostic impact demonstrated in previous studies. The following variables were entered into the multivariable model: age, diabetes, hypertension, hemoglobin levels at baseline, multivessel disease, ST-segment elevation myocardial infarction, decreased HDL, eGFR, TG levels at baseline, onset to balloon time, and use of drug-eluting stent. A 2-sided p value <0.05 was considered statistically significant.
Results
Baseline clinical characteristics at discharge between the 2 groups are listed in Table 1 . The decreased HDL group was older and had lower eGFR. The proportion of smoking cessation after AMI was similar between the 2 groups. There was no significant difference between the 2 groups in terms of medical treatment. The left anterior descending artery as the culprit vessel was more frequent in the increased HDL group; in contrast, the right coronary artery was more frequent in the decreased HDL group. The decreased HDL group had a higher rate of achievement of post–Thrombolysis In Myocardial Infarction 3, compared with the increased HDL group.
| Variable | High-Density Lipoprotein | P Value | |
|---|---|---|---|
| Increased (n = 620) | Decreased (n = 104) | ||
| Men | 505 (82%) | 89 (86%) | 0.19 |
| Age (years) | 61.8 ± 10.3 | 65.3 ± 9.0 | <0.01 |
| Body mass index (kg/m 2 ) | 24.2 ± 3.0 | 24.0 ± 3.3 | 0.44 |
| Diabetes mellitus | 148 (24%) | 22 (21%) | 0.32 |
| Hypertension | 304 (49%) | 56 (54%) | 0.21 |
| Dyslipidemia | 333 (54%) | 47 (45%) | 0.07 |
| Current smoker | 339 (55%) | 57 (55%) | 0.53 |
| Smoking cessation after AMI | 267 (79%) | 44 (77%) | 0.45 |
| Left ventricular ejection fraction (%) | 58.4 ± 10.6 | 59.8 ± 11.6 | 0.26 |
| Systolic blood pressure (mm Hg) | 139 ± 33 | 144 ± 38 | 0.33 |
| Diastolic blood pressure (mm Hg) | 85 ± 21 | 84 ± 22 | 0.36 |
| ST elevation AMI | 499 (81%) | 87 (85%) | 0.21 |
| Hemoglobin (g/dL) | 12.9 ± 1.8 | 12.8 ± 1.8 | 0.68 |
| Killip ≥2 on admission | 52 (8%) | 13 (13%) | 0.12 |
| Peak creatine kinase (mg/dL) | 2512 (1364 – 3946) | 2443 (1243 – 3581) | 0.31 |
| eGFR (ml/min/1.73 m 2 ) | 68.9 ± 15.8 | 63.1 ± 14.8 | <0.01 |
| Hemoglobin A1c (%) | 5.83 ± 1.44 | 5.77 ± 1.09 | 0.69 |
| Brain natriuretic peptide (pg/mL) | 103.2 (49.3 – 212.1) | 96.5 (47.8 – 160) | 0.54 |
| Infarct-related coronary artery | |||
| Right | 232 (37%) | 55 (53%) | <0.01 |
| Left anterior descending | 294 (47%) | 38 (37%) | 0.03 |
| Left circumflex | 92 (15%) | 11 (11%) | 0.16 |
| Left main | 2 (0.3%) | 0 (0%) | 0.73 |
| Multivessel coronary disease | 261 (42%) | 44 (42%) | 0.53 |
| Onset to balloon time (sec) | 225 (160 – 374) | 230 (152 – 485) | 0.37 |
| Post-TIMI 3 achievement | 590 (95%) | 104 (100%) | 0.02 |
| Drug eluting stent use | 234 (38%) | 47 (45%) | 0.09 |
Lipid profiles at baseline and follow-up at 6 to 9 months after AMI are listed in Table 2 . No significant differences were observed in terms of LDL-C and TG levels at baseline between the 2 groups. The decreased HDL group showed significantly higher HDL-C levels at baseline compared with the increased HDL group. As a result, the decreased HDL group showed significantly higher total cholesterol levels at baseline and lower LDL/HDL-C ratio compared with the increased HDL group. Both groups achieved follow-up LDL-C levels <100 mg/dl, which is currently a major target for secondary prevention after PCI. Patients in the decreased HDL group were more likely to have lower follow-up HDL-C levels and higher LDL/HDL-C ratio compared with those in the increased HDL group.
| Variable | High-Density Lipoprotein | P Value | |
|---|---|---|---|
| Increased (n = 620) | Decreased (n = 104) | ||
| Baseline | |||
| Total cholesterol (mg/dL) | 171 ± 36 | 181 ± 42 | 0.01 |
| TG (mg/dL) | 136 ± 56 | 129 ± 47 | 0.24 |
| LDL-C (mg/dL) | 124 ± 35 | 128 ± 35 | 0.36 |
| HDL-C (mg/dL) | 37 ± 9 | 44 ± 11 | <0.01 |
| LDL/HDL-C ratio | 3.6 ± 1.4 | 3.0 ± 1.0 | <0.01 |
| Follow-up | |||
| Follow-up period (days) | 236 ± 111 | 218 ± 61 | 0.18 |
| TG (mg/dL) | 142 ± 112 | 135 ± 76 | 0.54 |
| LDL-C (mg/dL) | 96 ± 26 | 93 ± 24 | 0.27 |
| HDL-C (mg/dL) | 48 ± 12 | 40 ± 9 | <0.01 |
| LDL/HDL-C ratio | 2.1 ± 0.7 | 2.5 ± 0.7 | <0.01 |
| ΔTG (mg/dL) | −7.0 (−38 to 28) | 2.0 (−37 to 44) | 0.29 |
| ΔLDL-C (mg/dL) | −28 (−49 to −6) | −32 (−53 to −17) | 0.07 |
| ΔHDL-C (mg/dL) | 10 (5 to 16) | −4.0 (−7.0 to −2.0) | <0.01 |
Mean follow-up periods were 1,471 ± 842 days. The cumulative cardiac event rate after the initial procedure is presented in Table 3 , and the event-free survival curves for adverse cardiovascular events are shown in the Figure 1 . There was no significant difference between the 2 groups in incidence of myocardial infarction, congestive heart failure, and target vessel revascularization. The incidence of all-cause death and stroke in the decreased HDL group was significantly higher than that in the increased HDL group. As a result, the rate of composite of adverse cardiovascular events in the decreased HDL group was numerically higher than that in the increased HDL group. Table 4 lists the multivariate predictors of adverse cardiovascular events. In this model, decreased HDL was the independent predictor of a composite of adverse cardiovascular events. In addition, onset to balloon time, multivessel disease, and hemoglobin levels at baseline independently predicted adverse cardiovascular events.
| Variable | High-Density Lipoprotein | P Value | |
|---|---|---|---|
| Increased (n = 620) | Decreased (n = 104) | ||
| Composite of all-cause death, recurrent myocardial infarction, and stroke | 44 (7%) | 16 (15%) | 0.01 |
| All-cause death | 18 (3%) | 8 (8%) | 0.02 |
| Cardiac death | 6 (1%) | 3 (3%) | 0.13 |
| Myocardial infarction | 22 (4%) | 3 (3%) | 0.51 |
| Stroke | 8 (1%) | 8 (8%) | <0.01 |
| Other | |||
| Target vessel revascularization | 81 (13%) | 15 (14%) | 0.40 |
| Congestive heart failure | 25 (4%) | 8 (8%) | 0.09 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
