Baseline characteristics of study subjects according to lipid-lowering strategies.
A total of 1275 patients were enrolled, the mean age was 62.3±10.4 years old, 72.8% was male, 39.1% (498) patients treated with intensive statins, and 60.9% (777) patients treated with conventional-dose statins plus ezetimibe. Baseline clinical characteristics and laboratory examinations indexes are summarized in Table 1.
Table 1
Baseline characteristics of study subjects according to lipid-lowering therapy.
Characteristics | Overall(n=1275) | Intensive statins(n=498) | Statins combined with ezetimibe(n=777) | P-value |
Demographics |
Age, years | 62.3±10.4 | 61.6±10.4 | 62.8±10.3 | 0.062 |
Sex, male (%) | 928 (72.8) | 380 (76.3) | 548 (70.5) | 0.028 |
BMI, kg/m2 | 24.7±2.8 | 24.6±2.8 | 24.8±2.8 | 0.230 |
Current smoking (%) | 349 (27.4) | 167 (33.5) | 182 (23.4) | <0.001 |
Laboratory examination |
HbA1c (%) | 5.9 (5.5, 6.7) | 6.0 (5.5, 6.8) | 5.9 (5.5, 6.6) | 0.835 |
eGFR (ml/min/1.73m2) | 90.9 (78.9, 99.2) | 91.2 (78.2, 100.2) | 90.5 (79.6, 99.0) | 0.408 |
CRP mean, mg/L | 1.52 (0.82, 2.73) | 1.74 (0.89, 3.00) | 1.42 (0.78, 2.62) | 0.007 |
NLR mean | 3.0 (2.3, 4.0) | 3.1 (2.4, 4.1) | 3.0 (2.3, 3.9) | 0.069 |
TC mean, mmol/L | 3.8 (3.3, 4.3) | 3.6 (3.2, 4.2) | 3.9 (3.4, 4.4) | <0.001 |
TC SD | 790.3(492.2, 1189.8) | 657.6 (397.0, 982.6) | 872.5 (586.7, 1281.2) | <0.001 |
TC CV | 207.6 (135.0, 300.3) | 182.9 (115.8, 249.1) | 228.3 (150.0, 316.6) | <0.001 |
TC VIM | 9.4 (5.0, 16.2) | 7.8 (4.0, 12.3) | 10.8 (5.8, 17.5) | <0.001 |
TG mean, mmol/L | 1.4 (1.1, 1.8) | 1.4 (1.1, 1.8) | 1.4 (1.1, 1.9) | 0.139 |
TG SD | 363.6 (215.2, 564.4) | 342.0 (211.0, 536.2) | 369.8 (216.0, 581.9) | 0.304 |
TG CV | 263.4 (177.8, 369.7) | 265.3 (181.2, 369.0) | 262.8 (175.1, 370.1) | 0.985 |
TG VIM | 23.5 (12.5, 40.6) | 23.8 (12.9, 40.5) | 23.4 (12.2, 40.7) | 0.985 |
LDL-C mean, mmol/L | 1.9 (1.6, 2.3) | 1.8 (1.5, 2.2) | 1.9 (1.6, 2.3) | <0.001 |
LDL-C SD | 655.3 (413.6, 958.1) | 540.5 (310.3, 836.0) | 719.6 (495.4, 1018.8) | <0.001 |
LDL-C CV | 343.8 (224.8, 470.4) | 291.7 (187.9, 416.6) | 375.9 (259.4, 501.9) | <0.001 |
LDL-C VIM | 92.2 (59.1, 127.9) | 77.6 (49.0, 112.7) | 101.2 (68.6, 136.9) | <0.001 |
Non-HDL-C mean, mmol/L | 2.5 (2.2, 3.0) | 2.5 (2.1, 2.9) | 2.6 (2.2, 3.1) | <0.001 |
Non-HDL-C SD | 526.6 (327.6, 819.9) | 455.5 (277.0, 685.5) | 601.7 (374.4, 922.9) | <0.001 |
Non-HDL-C CV | 204.7 (136.4, 297.1) | 180.0 (116.90, 254.0) | 220.3 (149.7, 328.7) | <0.001 |
Non-HDL-C VIM | 20.7 (12.4, 33.1) | 17.6 (10.2, 27.2) | 22.7 (13.9, 37.6) | <0.001 |
Comorbidities |
Hypertension (%) | 796 (62.43) | 319 (64.1) | 477 (61.4) | 0.368 |
Diabetes mellitus (%) | 309 (24.24) | 139 (27.9) | 170 (21.9) | 0.017 |
Previous PCI (%) | 60 (7.59) | 27 (6.4) | 33 (9.0) | 0.209 |
Previous MI (%) | 8 (1.07) | 18 (4.3) | 13 (3.5) | 0.746 |
Medication at discharge |
Aspirin (%) | 1240 (97.25) | 485 (97.4) | 755 (97.2) | 0.952 |
ACEI (%) | 340 (26.67) | 179 (35.9) | 161 (20.7) | <0.001 |
ARB (%) | 455 (35.69) | 156 (31.3) | 299 (38.5) | 0.011 |
Βeta-blocker (%) | 773 (60.63) | 339 (68.1) | 434 (55.9) | <0.001 |
CCB (%) | 329 (25.80) | 108 (21.7) | 221 (28.4) | 0.009 |
Data were expressed as mean ± SD, or number (%). BMI=body mass index, eGFR=estimated glomerular filtration rate, CRP=C-reactive protein, NLR=neutrophil to lymphocyte ratio, TC=total cholesterol, TG=triglyceride, LDL-C=low-density lipoprotein cholesterol, non-HDL-C=non-high-density lipoprotein cholesterol, SD=standard deviation, CV=coefficient of variation, VIM=variability independent of mean, PCI=percutaneous coronary intervention, MI=myocardial infarction, ACEI=angiotensin converting enzyme inhibitor, ARB=angiotensin receptor antagonists, CCB=calcium channel blocker. |
Compared with conventional-dose statins combined with ezetimibe group, patients in intensive statins group were more males (76.3% vs. 70.5%, P=0.028), more smokers (33.5% vs. 23.4%, P <0.001), greater burden of diabetes mellitus (27.9% vs 21.9%, P= 0.017) and were more likely to be treated with beta-blocker, calcium channel blocker (CCB), angiotensin converting enzyme inhibitor (ACEI), and angiotensin receptor antagonists (ARB) (p for all <0.05). In addition, patients in intensive statins group had lower variability in plasma TC (7.8 vs. 10.8 mmol/L, P <0.001), LDL-C (77.6 vs. 100.9 mmol/L, P <0.001), and non-HDL-C (17.6 vs. 22.7 mmol/L, P <0.001) levels by VIM method. However, there was no significant difference in the variability in plasma TG levels between two groups. The similar results were found in CV and SD method.
Multiple linear regression analysis of lipid-lowering strategies to variability in plasma lipids levels in patients with CAD.
In order to accurately evaluate the effect of lipid-lowering strategies on lipid stability, linear regression was performed to estimate the relationship between lipid-lowering strategies and variability of plasma lipids levels in patients with CAD. (Table 2)
Table 2
Coefficients of combination therapy (vs. intensive therapy) for the variability of TC, TG, and LDL, non-HDL cholesterol in linear regression analysis (95% confidence interval).
| Model 1a | | Model 2b | | Model 3c |
| β coefficient(95% CI) | P value | β coefficient(95% CI) | P value | β coefficient(95% CI) | P value |
TC variability | | | | | | |
VIM | 3.6(2.7 to 4.5) | <0.001 | 3.8(2.8 to 4.7) | <0.001 | 3.6(2.7 to 4.6) | <0.001 |
CV | 49.7(37.0 to 62.4) | <0.001 | 52.0(39.0 to 65.1) | <0.001 | 50.1(37.0 to 63.3) | <0.001 |
SD* | 162.9(109.5 to 216.3) | <0.001 | 179.5(125.0 to 234.0) | <0.001 | 173.0(118.1 to 227.8) | <0.001 |
TG variability | | | | | | |
VIM | 0.7(-3.3 to 4.6) | 0.742 | 1.7(-2.4 to 5.9) | 0.411 | 1.0(-3.2 to 5.1) | 0.655 |
CV | 2.9(-16.8 to 22.7) | 0.771 | 9.3(-11.2 to 29.9) | 0.373 | 5.7(-14.9 to 26.4) | 0.585 |
SD* | -2.1(-62.0 to 57.7) | 0.944 | 6.2(-57.3 to 69.7) | 0.849 | -2.9(-66.8 to 60.9) | 0.928 |
LDL-C variability | | | | | | |
VIM | 23.6(18.1 to 29.0) | <0.001 | 24.2(18.6 to 30.0) | <0.001 | 23.6(17.9 to 29.3) | <0.001 |
CV | 83.9(64.4 to 103.4) | <0.001 | 86.4(66.2 to 106.6) | <0.001 | 84.1(63.8 to 104.3) | <0.001 |
SD* | 149.2(108.6 to 189.8) | <0.001 | 155.5(113.8 to 197.1) | <0.001 | 148.3(107.5 to 189.1) | <0.001 |
Non-HDL-C variability | | | | | | |
VIM | 8.0(5.8 to 10.1) | <0.001 | 8.2(5.9 to 10.4) | <0.001 | 8.2(5.9 to 10.4) | <0.001 |
CV | 56.7(41.4 to 72.0) | <0.001 | 58.0(42.0 to 74.0) | <0.001 | 58.3(42.2 to 74.4) | <0.001 |
SD* | 139.7(97.2 to 182.2) | <0.001 | 148.6(104.3 to 192.8) | <0.001 | 147.7(103.2 to 192.3) | <0.001 |
CI=confidence interval, CV=coefficient of variation; VIM=variability independent of mean; SD=standard deviation; |
Model 1a: Adjusted for none. |
Model 2b: Adjusted for age, sex, body mass index, current smoking, diabetes mellitus, hypertension, mean NLR level. |
Model 3c: Additionally adjusted for medications (aspirin, ACEI or ARB, Βeta-blocker, CCB). |
SD*: Additionally adjusted for the mean of corresponding plasma lipids variability. |
P<0.05 was in bold. |
TC, TG, LDL-C, non-HDL-C were the same as Table 1. |
CHD=coronary heart disease, LVEF=left ventricular ejection fraction, NYHA=New York Heart Association, eGFR=estimated glomerular filtration rate, ALT=alanine aminotransferase. |
TC, LDL-C,and non-HDL-C were the same as Figure 2. |
TC, LDL-C,and non-HDL-C were the same as Figure 2. |
Without any multi-variable adjusted, Model 1 showed conventional-dose statins combined with ezetimibe group had higher variability in TC levels in SD, CV and VIM methods (SD: β = 162.9, 95% confidence interval [CI] [109.5, 216.3], p<0.001; CV: β = 49.7, 95%CI [37.0, 62.4], p<0.001; VIM: β = 3.6, 95%CI [2.7, 4.5], p<0.001), higher variability in LDL-C levels (SD: β = 149.2, 95%CI [108.6, 189.8], p<0.001; CV: β = 83.9, 95%CI [64.4, 103.4], p<0.001; VIM: β = 23.6, 95%CI [18.1, 29.0], p<0.001), and higher variability in Non-HDL-C levels (SD: β = 139.7, 95%CI [97.2, 182.2], p<0.001; CV: β = 56.7, 95%CI [41.4, 72.0], p<0.001; VIM: β = 8.0, 95%CI [5.8, 10.1], p<0.001). No significant difference in TG variability were found between two groups.
After adjusted for basic characteristics (including age, sex, body mass index, current smoking, diabetes mellitus, hypertension, and mean neutrophil to lymphocyte ratio (NLR) level), the similar results were confirmed in Model 2. After further adjusted for other medications (including aspirin, ACEI or ARB, Βeta-blocker, and CCB), the statins plus ezetimibe therapy had higher lipids variability than intensive statins in patients with CAD in Model 3: higher TC variability (SD: β = 173.0, 95%CI [118.1, 227.8], p<0.001; CV: β = 50.1, 95%CI [37.0, 63.3], p<0.001; VIM: β = 3.6, 95%CI [2.7, 4.6], p<0.001), LDL-C variability (SD: β = 148.3, 95%CI [107.5, 189.1], p<0.001; CV: β = 84.1, 95%CI [63.8, 104.3], p<0.001; VIM: β = 23.6, 95%CI [17.9, 29.3], p<0.001), and Non-HDL-C variability (SD: β = 147.7, 95%CI [103.2, 192.3], p<0.001; CV: β = 58.3, 95%CI [42.2, 74.4], p<0.001; VIM: β = 8.2, 95%CI [5.9, 10.4], p<0.001).
Linear regression analysis of lipid-lowering strategies to variability in plasma lipids levels in the subgroups.
To investigate whether the results were stable, subgroup analysis based on atorvastatin or rosuvastatin, diabetes mellites or not, hypertension or not (Figure 4), also suggested that intensive statins group had lower variability in TC, LDL-C and non-HDL-C levels in VIM method compare with conventional-dose statins combined with ezetimibe group (p for all<0.05). The results were like those of the primary analysis when variability of plasma lipids was represented by SD (Figure 2) or CV (Figure 3).