Lipoprotein (a) variability is associated with mean follow-up C-reactive protein in patients with coronary artery disease following percutaneous coronary intervention

Abstract

BACKGROUND

Increased lipoprotein (a) [lp (a)] has proinflammatory effects, which increase the risk of coronary artery disease. However, the association between lp (a) variability and follow-up C-reactive protein (CRP) level in patients undergoing percutaneous coronary intervention (PCI) has not been investigated.

AIM

To explore the association between lp (a) variability and mean CRP levels within the 1^st year post-PCI.

METHODS

Results of lp (a) and CRP measurements from at least three follow-up visits of patients who had received PCI were retrospectively analyzed. Standard deviation (SD), coefficient of variation (CV), and variability independent of the mean (VIM) are presented for the variability for lp (a) and linear regression analysis was conducted to correlate lp (a) variability and mean follow-up CRP level. The relationship of lp (a) variability and inflammation status was analyzed by restricted cubic spline analysis. Finally, exploratory analysis was performed to test the consistency of results in different populations.

RESULTS

A total of 2712 patients were enrolled. Patients with higher variability of lp (a) had a higher level of mean follow-up CRP (P < 0.001). lp (a) variability was positively correlated with the mean follow-up CRP (SD: β = 0.023, P < 0.001; CV: β = 0.929, P < 0.001; VIM: β = 1.648, P < 0.001) by multivariable linear regression analysis. Exploratory analysis showed that the positive association remained consistent in most subpopulations.

CONCLUSION

Lp (a) variability correlated with mean follow-up CRP level and high variability could be considered an independent risk factor for increased post-PCI CRP level.

Key Words: Lipoprotein (a), Variability, C-reactive protein, Coronary artery disease, Percutaneous coronary intervention

Core Tip: In this study, we determined the predictive value of lipoprotein (a) [lp (a)] variability for the C-reactive protein (CRP) level during the 1-year of follow-up after percutaneous coronary intervention (PCI). One of the main strengths is the large sample size of the study. Additionally we used multiple measures to methods to validate our results. In this multicenter retrospective study, the variability of lp (a) was closely correlated with the average follow-up CRP levels, and high lp (a) variability could be considered an independent risk factor for increased CRP levels during follow-up visits for patients treated with PCI.

INTRODUCTION

Coronary artery disease (CAD) is a major global public health problem, and percutaneous coronary intervention (PCI) to reconstruct coronary blood flow and improve prognosis has been of great benefit[1,2]. However, stent deformation, acute renal injury, stent thrombosis, and target lesion revascularization often occur post-PCI and lead to poor prognosis[3-5].

Elevated inflammatory biomarkers have been linked to a poor prognosis post-PCI and C-reactive protein (CRP), a critical indicator of inflammation status, is a predictor of cardiovascular disease risk[6,7]. CRP influences several pathogenic pathways associated with atherosclerosis and promotes the adhesion between monocytes and endothelial cells, low-density lipoprotein (LDL) accumulation in macrophages, platelet aggregation, and production of reactive oxygen species and reduces endothelial cell nitric oxide formation[8]. Such pathological mechanisms combine to induce arterial inflammation, formation of foam cells, and thrombosis[9-11]. Therefore, elevated CRP is considered an independent risk factor for cardiovascular events following PCI[12].

Atherosclerosis is a chronic disease characterized by disordered lipid metabolism and chronic inflammation, which are considered to be co-dependent processes. Indeed, aberrant lipid metabolism promotes both atherosclerotic progression and nonbacterial inflammatory reactions[13,14]. The lipid profile has also been closely related to coronary artery disease (CAD) prognosis[15].

Lipoprotein a [lp (a)] is an atherogenic lipoprotein, similar in structure to LDL cholesterol (LDL-C), which consists of an LDL-like particle and apolipoprotein B-100 (apo B100)[16]. By contrast with LDL-C, circulating lp (a) levels are primarily determined by activity of the LPA gene, without significant dietary or environmental influence, mediating CAD risk throughout a patient’s lifetime[17]. For these reasons, lp (a) variability has not received much attention but therapeutic agents targeting lp (a) are recently progressing through randomized clinical trials, such as proprotein convertase subtilisin/kexin type 9 serine protease inhibitors, IONIS-APO (a)Rx, etc, which makes lp (a) lowering therapy possible[18-20]. Circulating lp (a) levels are not as stable as they used to be. Higher lp (a) levels have been associated with elevated CRP, producing a greater risk of atherosclerotic cardiovascular disease[21]. However, neither the relationship between lp (a) and CRP in PCI-treated patients nor the association between lp (a) variability and CRP during PCI follow-up examinations has been explored.

The association between lp (a) variability and mean CRP level among patients participating in at least three follow-up visits within the 1^st year post-PCI was examined during the current multicenter retrospective study.

MATERIALS AND METHODS

Study population

Data from patients treated with PCI between January 2010 and January 2019 at Sir Run Run Shaw Hospital and its medical consortium hospitals were retrospectively collated. Inclusion criteria were as follows: complete baseline data; at least three laboratory measurements within 1-year follow-up; and CRP levels elevated by no more than 10 mg/dL compared to baseline. Exclusion criteria were as follows: acute or chronic inflammatory status, concomitant heart failure (New York Heart Association Grade III-IV) or valvular diseases, severe hepatic or renal dysfunction, and active malignant tumor. All participants provided informed consent for use of data in scientific research and ethical approval was granted by the Ethics Review Committee of Sir Run Run Shaw Hospital (No. 20201217-36). The study was performed in accordance with the 1975 Declaration of Helsinki.

Data collection

Demographic data (age, sex, height, weight, smoking status, and past medical history), laboratory measurements and medications at discharge for all eligible patients were collected from the Hospital Information System. Each patient in this study was asked to participate in at least three follow-up examinations within the 1^st year after PCI. Routine biochemical analysis of fasting blood samples was performed by the local clinical laboratory at each of three follow-up visits. Total serum cholesterol, high-density lipoprotein cholesterol (HDL-C) and triglyceride were measured enzymatically and the Friedewald equation was applied to estimate LDL-C concentrations. lp (a) measurements were performed using immunoturbidimetric assay (Denka Seiken Co. Ltd., Tokyo, Japan) and CRP by transmission turbidimetry on an automatic analyzer (Advia 2400; Siemens, Munich, Germany). White blood cell counts were measured by an automatic blood cell counter (XE-2100; Sysmex, Kobe, Japan).

Variability definitions

Lp (a) variability was calculated and defined as the standard deviation (SD), coefficient of variation (CV), and variability independent of the mean (VIM) during follow-up visits. Calculations were performed as follows: SD: arithmetic square root of the square of the difference between measurements and the mean; CV: SD/mean × 100%; VIM: SD/mean^β × 100%, in which β is the regression coefficient based on the natural logarithm of the SD and mean.

Statistical analysis

Categorical variables are presented as numbers and percentages, and intergroup comparisons carried out using the Chi-square test. Continuous non-normally distributed variables are expressed as the median and interquartile range and comparisons were conducted with the Mann-Whitney U test. The patient cohort was divided into two groups based on the median value of lp (a) CV and baseline characteristics were compared. Potential risk factors linked to mean follow-up CRP level were identified by univariable linear regression between the baseline features and laboratory parameters. Factors with P < 0.10 were used for stepwise multivariable linear regression. A similar analysis was conducted for all three measurements of lp (a) variability. Restricted cubic spline (RCS) analysis of lp (a) variability and high inflammatory status using the median follow-up CRP level of 1.52 mg/L was performed. Exploratory analysis was conducted to determine whether the association between lp (a) variability and mean follow-up CRP was consistent across all subgroups. Subgroups were stratified by age, sex, diabetes, hypertension, and mean follow-up lp (a) concentration. P < 0.05 was considered statistically significant unless otherwise indicated. Statistical analyses were performed using SPSS 25.0 software (Chicago, IL, United States) and R 4.0.5 software (Vienna, Austria).

RESULTS

Baseline characteristics and laboratory measurements

Baseline characteristics and follow-up laboratory measurements for the 2712 patients enrolled in the current study are shown in Table 1. Median age was 64.0 years and 72.2% were male. There was a tendency for patients with a high follow-up lp (a) CV to have diabetes but no intergroup differences in hypertension or previous PCI history were found. There were differences in mean HDL-C, triglyceride, CRP and lp (a) but none in total cholesterol, LDL-C or white blood cell (WBC). Interestingly, patients who had higher lp (a) CV had lower follow-up lp (a) but higher follow-up CRP (P < 0.001). Grouping by lp (a) SD or VIM shown in the Supplementary Tables 1 and 2, variables with between-group differences were similar to grouping by lp (a) CV (Figure 1).

Figure 1 Study flowchart. CRP: C-reactive protein; CV: Coefficient of variation; NYHA: New York Heart Association; PCI: Percutaneous coronary intervention.

Table 1 Baseline characteristics.

		Follow-up lp (a) CV
Parameter	Overall, n = 2712	< 0.22, n = 1388	≥ 0.22, n = 1324	P value
Demographics
Age, yr	64.0 (58.0, 72.0)	64.0 (58.0, 72.0)	64.0 (57.0, 72.0)	0.593
Male, n (%)	1957 (72.2)	982 (70.7)	975 (73.6)	0.102
BMI, kg/m2	24.7 (22.7, 26.1)	24.6 (22.6, 26.1)	24.7 (22.8, 26.2)	0.191
Current smoking, n (%)	710 (26.2)	337 (24.3)	373 (28.2)	0.024a
Diabetes, n (%)	692 (25.5)	327 (23.6)	365 (27.6)	0.019a
Hypertension, n (%)	1727 (63.7)	869 (62.6)	858 (64.8)	0.251
Previous PCI, n (%)	172 (6.3)	92 (6.6)	80 (6.0)	0.584
Laboratory measurements
Average TC level, mmol/L	3.67 (3.23, 4.18)	3.67 (3.24, 4.16)	3.67 (3.21, 4.19)	0.737
Average LDL-C level, mmol/L	1.80 (1.50, 2.18)	1.80 (1.49, 2.17)	1.79 (1.50, 2.18)	0.968
Average HDL-C level, mmol/L	1.00 (0.86, 1.18)	1.03 (0.88, 1.20)	0.98 (0.84, 1.15)	< 0.001a
Average TG level, mmol/L	1.34 (1.04, 1.83)	1.32 (1.02, 1.75)	1.37 (1.06, 1.91)	0.002a
Average CRP level, mg/L	1.52 (0.80, 2.83)	1.40 (0.72, 2.60)	1.65 (0.87, 3.00)	< 0.001a
Average WBC level, × 109/L	6.38 (5.51, 7.25)	6.30 (5.46, 7.25)	6.43 (5.55, 7.26)	0.089
Average lp (a) level, mg/dL	14.00 (7.32, 34.90)	16.70 (7.16, 45.81)	12.52 (7.36, 25.48)	< 0.001a
Lp (a) SD	3.38 (1.61, 7.49)	2.19 (0.95, 5.40)	4.55 (2.61, 9.16)	< 0.001a
Lp (a) CV	0.22 (0.14, 0.32)	0.14 (0.10, 0.18)	0.32 (0.26, 0.43)	< 0.001a
Lp (a) VIM	0.17 (0.13, 0.23)	0.13 (0.10, 0.15)	0.24 (0.20, 0.29)	< 0.001a
Medication, n (%)
Statin	2675 (98.6)	1379 (99.4)	1296 (97.9)	0.002a
Ezetimibe	451 (16.6)	218 (15.7)	233 (17.6)	0.204
ACEI or ARB	1588 (58.6)	776 (55.9)	812 (61.3)	0.005a
Beta blocker	1639 (60.4)	823 (59.3)	816 (61.6)	0.228

^aP < 0.05. Data are presented as the median (interquartile range) or numbers (proportions). ACEI: Angiotensin-converting enzyme inhibitor; ARB: Angiotensin receptor antagonist; BMI: Body mass index; CRP: C-reactive protein; CV: Coefficient of variation; HDL-C: High-density lipoprotein cholesterol; LDL-C: Low-density lipoprotein cholesterol; Lp (a): indicates lipoprotein (a); PCI: Percutaneous coronary intervention; SD: Standard deviation; TC: Total cholesterol; TG: Triglyceride; VIM: Variability independent of the mean; WBC: White blood cell.

Lp (a) variability and inflammatory status

A follow-up CRP level higher than the median of 1.52 mg/L was considered to show high inflammatory status. RCS analysis suggested that the risk of hyperinflammation gradually increased with increasing lp (a) SD and then flattened out (Figure 2A) and initial shallow increases in lp (a) CV and VIM were followed by a rapid rate of increase (Figure 2B and C). A non-linear association was observed between lp (a) SD and hyperinflammation but linear trends between lp (a) CV or VIM and hyperinflammation (SD: P for nonlinearity< 0.001; CV: P for nonlinearity = 0.4653; VIM: P for nonlinearity = 0.4344).

Figure 2 Restricted cubic spline analysis of lipoprotein (a) variability with relatively high inflammation status. Patients with a relatively high inflammation status were defined as those whose average follow-up C-reactive protein was more than the median level (1.52 mg/L). The spline curves were plotted and P for nonlinearity was calculated in each variability of lipoprotein (a) [lp (a)]. A: The risk of hyperinflammation gradually increased with increasing lp (a) SD and then flattened out; B and C: Initial shallow increases in lp (a) coefficient of variation and variability independent of the mean were followed by a rapid rate of increase. CI: Confidence interval; CRP: C-reactive protein; CV: Coefficient of variation; Lp (a): Lipoprotein (a); SD: Standard deviation; VIM: Variability independent of the mean.

Lp (a) variability and follow-up CRP level

Univariable linear regression of lp (a) variability at follow-up and other potential risk factors was performed to assess the impact on mean follow-up CRP level (Tables 2-4). The SD, CV, and VIM of lp (a) variability showed a positive relationship with mean follow-up CRP level (SD: β = 0.020, P < 0.001; CV: β = 0.961, P < 0.001; VIM: β = 1.692, P < 0.001). The positive relationship persisted after regression analysis of mean lp (a) level, age, diabetes, hypertension, BMI, average LDL-C level, average WBC count, statin and ezetimibe, regardless of the values of lp (a) SD, CV or VIM (SD: β = 0.023, P < 0.001; CV: β = 0.929, P < 0.001; VIM: β = 1.648, P < 0.001). In addition, age, diabetes, BMI, mean LDL-C level and mean WBC count were all independent risk factors for mean follow-up CRP while statin and ezetimibe were independent protective factors.

Table 2 Association of lp (a) standard deviation with average C-reactive protein level in linear regression.

Parameter	Univariable analysisb		Multivariable analysis
	β coefficient (95%CI)	P value	β coefficient (95%CI)	P value
Lp (a) SD	0.020 (0.010, 0.029)	< 0.001	0.023 (0.009, 0.037)	0.001a
Average lp (a) level, mg/dL	0.002 (0.000, 0.004)	0.059	-0.002 (-0.005, 0.001)	0.183
Age in yr	0.026 (0.020, 0.031)	< 0.001	0.030 (0.025, 0.036)	< 0.001a
Male	-0.017 (-0.152, 0.119)	0.811
Diabetes	0.312 (0.173, 0.451)	< 0.001	0.201 (0.062, 0.340)	0.005a
Hypertension	0.314 (0.188, 0.440)	< 0.001	0.133 (0.003, 0.263)	0.045a
BMI, kg/m2	0.029 (0.009, 0.049)	0.004	0.028 (0.008, 0.048)	0.007a
Current smoking	0.104 (-0.034, 0.242)	0.139
Average LDL-C level, mmol/L	0.215 (0.117, 0.312)	< 0.001	0.200 (0.098, 0.301)	< 0.001a
Average WBC count, × 109/L	0.256 (0.207, 0.304)	< 0.001	0.264 (0.216, 0.312)	< 0.001a
Previous PCI	0.041 (-0.208, 0.291)	0.744
Statin	-0.662 (-1.186, 0.139)	-0.013	-0.706 (-1.229, 0.183)	-0.008a
Ezetimibe	-0.202 (-0.365, 0.039)	-0.015	-0.334 (-0.497, -0.170)	< 0.001a

^aP < 0.05.

^bPredictors with P < 0.10 in univariable linear regression analysis were entered into the subsequent multivariable linear regression analysis. BMI: Body mass index; CI: Confidence interval; CRP: C-reactive protein; LDL-C: Low-density lipoprotein cholesterol; Lp (a): Lipoprotein (a); PCI: Percutaneous coronary intervention; SD: Standard deviation; WBC: White blood cell.

Table 3 Association of lp (a) coefficient of variation with average C-reactive protein level in linear regression.

	Univariable analysisb		Multivariable analysis
	β coefficient (95%CI)	P value	β coefficient (95%CI)	P value
Lp (a) CV	0.961 (0.588, 1.334)	< 0.001	0.929 (0.554, 1.304)	< 0.001a
Average lp (a) level, mg/dL	0.002 (0.000, 0.004)	0.059	0.003 (0.001, 0.005)	0.013a
Age in yr	0.026 (0.020, 0.031)	< 0.001	0.031 (0.025, 0.036)	< 0.001a
Male	-0.017 (-0.152, 0.119)	0.811
Diabetes	0.312 (0.173, 0.451)	< 0.001	0.201 (0.062, 0.340)	0.005a
Hypertension	0.314 (0.188, 0.440)	< 0.001	0.123 (-0.007, 0.253)	0.064
BMI, kg/m2	0.029 (0.009, 0.049)	0.004	0.028 (0.008, 0.048)	0.006a
Current smoking	0.104 (-0.034, 0.242)	0.139
Average LDL-C level, mmol/L	0.215 (0.117, 0.312)	< 0.001	0.200 (0.099, 0.301)	< 0.001a
Average WBC count, × 109/L	0.256 (0.207, 0.304)	< 0.001	0.261 (0.214, 0.309)	< 0.001a
Previous PCI	0.041 (-0.208, 0.291)	0.744
Statin	-0.662 (-1.186, 0.139)	-0.013	-0.653 (-1.175, 0.130)	-0.014a
Ezetimibe	-0.202 (-0.365, 0.039)	-0.015	-0.329 (-0.493, -0.166)	< 0.001a

^aP < 0.05.

^bPredictors with P < 0.10 in univariable linear regression analysis were entered into the subsequent multivariable linear regression analysis. BMI: Body mass index; CI: Confidence interval; CRP: C-reactive protein; CV: Coefficient of variation; LDL-C: Low-density lipoprotein cholesterol; Lp (a): Lipoprotein (a); PCI: Percutaneous coronary intervention; WBC: White blood cell.

Table 4 Association of lp (a) variability independent of the mean with average C-reactive protein level in linear regression.

	Univariable analysisb		Multivariable analysis
	β coefficient (95%CI)	P value	β coefficient (95%CI)	P value
Lp (a) VIM	1.692 (1.026, 2.359)	< 0.001	1.648 (0.977, 2.320)	< 0.001a
Average lp (a) level, mg/dL	0.002 (0.000, 0.004)	0.059	0.003 (0.001, 0.005)	0.013a
Age in yr	0.026 (0.020, 0.031)	< 0.001	0.031 (0.025, 0.037)	< 0.001a
Male	-0.017 (-0.152, 0.119)	0.811
Diabetes	0.312 (0.173, 0.451)	< 0.001	0.201 (0.062, 0.340)	0.005a
Hypertension	0.314 (0.188, 0.440)	< 0.001	0.124 (-0.006, 0.254)	0.062
BMI, kg/m2	0.029 (0.009, 0.049)	0.004	0.028 (0.008, 0.048)	0.006a
Current smoking	0.104 (-0.034, 0.242)	0.139
Average LDL-C level, mmol/L	0.215 (0.117, 0.312)	< 0.001	0.200 (0.099, 0.301)	< 0.001a
Average WBC count, × 109/L	0.256 (0.207, 0.304)	< 0.001	0.261 (0.213, 0.309)	< 0.001a
Previous PCI	0.041 (-0.208, 0.291)	0.744
Statin	-0.662 (-1.186, 0.139)	-0.013	-0.655 (-1.177, 0.132)	-0.014a
Ezetimibe	-0.202 (-0.365, 0.039)	-0.015	-0.331 (-0.494, -0.167)	< 0.001a

^aP < 0.05.

^bPredictors with P < 0.10 in univariable linear regression analysis were entered into the subsequent multivariable linear regression analysis. BMI: Body mass index; CI: Confidence interval; CRP: C-reactive protein; LDL-C: Low density lipoprotein cholesterol; Lp (a): Lipoprotein (a); PCI: Percutaneous coronary intervention; VIM: Variability independent of the mean; WBC: White blood cell.

Exploratory analysis

Exploratory analysis was conducted on the positive association between lp (a) variability and CRP among patients from different subgroups. Patients were divided into two groups according to age (< 65 years or ≥ 65 years), sex (female or male), presence of diabetes (yes or no), presence of hypertension (yes or no) and mean follow-up lp (a) levels (< 30 or ≥ 30 mg/dL; Figure 3). Multivariable linear regression was conducted using the same covariates from Tables 2-4. Significant positive associations were found for some subgroups regardless of age, presence of diabetes or presence of hypertension (P < 0.05). In particular, when studying the population with average follow-up lp (a) levels ≥ 30 mg/dL, the positive association no longer became significant in all measurements of variability of lp (a). Similar insignificant results were also found in female (in addition to lp (a) CV).

Figure 3 Exploratory analysis in various subgroups. The associations between Lp (a) variability and average follow-up C-reactive protein level were evaluated in various populations. The population was subdivided according to the age, sex, presence of diabetes, presence of hypertension, and average follow-up Lp (a) level. Multivariable linear regression analysis was repeated in each subgroup with the same adjustment as Tables 2-4. A: Mean follow-up lp (a) SD levels; B: Mean follow-up lp (a) CV levels; C: Mean follow-up lp (a) VIM levels. CI: Confidence interval; CRP: C-reactive protein; CV: Coefficient of variation; Lp (a): Lipoprotein (a); SD: Standard deviation; VIM: Variability independent of the mean.

DISCUSSION

This study investigated correlations between lp (a) and CRP during 1-year follow-up in CAD patients who underwent PCI and found a significant relationship between metrics of lp (a) variability, SD, CV, VIM, and mean CRP levels. High lp (a) variability was found to be an independent risk factor for increased CRP and high inflammatory status. Interestingly, the relationship was not significant in the subgroup with lp (a) ≥ 30 mg/dL, perhaps due to the lipid-lowering medication prescribed.

The proatherogenic lipoprotein, lp (a), has been regarded as a promising biomarker for cardiovascular disease risk[22]. Lp (a) activates monocytes to cause accumulation of cholesterol in atherosclerotic plaques, inducing migration and proliferation of vascular smooth muscle cells and stimulating the secretion of endothelial adhesion molecules[23]. These pathological changes accelerate subendothelial foam cell formation. lp (a) has an additional thrombogenic action due to its content of apo (a), which competes with plasminogen receptors for binding sites[24]. Oxidation of LDL and LDL receptor 1 results in endothelial dysfunction, accelerating the pathogenesis of atherosclerosis[25]. Lp (a) is a LDL particle with the glycoprotein, apo (a), covalently bound to apo B-100 and has also been implicated in atheroma development due to its association with the oxidizable properties of LDLs[25]. Lp (a) exerts antifibrinolytic effects and increases platelet aggregation around the plaque, contributing to the pathogenesis and development of atherosclerosis[26,27].

Individual lp (a) expression is usually considered to be stable and 90% is determined by genetic phenotype. However, variable expression is observed in specific pathological conditions involving inflammation status such as renal dysfunction and diabetes mellitus[28-30]. Zhang et al[31] reported that lp (a) expression increased two-fold above the normal range in response to severe oxidative stress and it is acknowledged that lp (a) levels respond to changes in the internal environment, especially in inflammatory status. Recent studies have focused on the relationship between lp (a) levels and the mild chronic inflammatory status observed in patients with atherosclerosis[32]. Garrafa et al[33] found significantly increased levels of lp (a) and CRP in older persons, suggesting a relationship between the two biomarkers.

Visit-to-visit lipid profile variability has received increasing clinical attention as an independent risk factor for adverse cardiovascular events[34]. Kim et al[15] demonstrated that follow-up cholesterol variability is an independent predictor for the incidence of cardiovascular events and Zhao et al[35] found that blood lipid variability increased systemic inflammation. However, no previous study has focused on the impact of lp (a) variability on CRP level, acknowledged to be an inflammation indicator, in patients treated with PCI. The current study revealed that mean follow-up CRP was associated with lp (a) variability in post-PCI patients. Multivariate regression analysis confirmed high lp (a) variability as a risk factor for elevated mean follow-up CRP. This study suggests that lp (a) variability is a reliable indicator of inflammatory status during post-PCI follow-up, the knowledge of which may guide clinicians to prevent recurrence of cardiovascular events.

Most subgroups of the current cohort, including age < 65 years or ≥ 65 years, male, diabetes or not, hypertension or not and lp (a) < 30 mg/dL, gave results consistent with the main findings. However, no relationship emerged between lp (a) variability and CRP levels in the lp (a) ≥ 30 mg/dL subgroup. Lipid-lowering drugs, such as statins and ezetimibe, also have anti-inflammatory actions[36]. The lipid-lowering therapy which is the conventional treatment for post-PCI CAD patients with lp (a) > 30 mg/dL, who tend to have higher plasma lipid levels may have the indirect effect of decreasing CRP levels.

Limitations

This study had some limitations. First, this was a multicenter retrospective observational study and additional prospective studies are required to verify the current findings. Second, other inflammatory markers such as procalcitonin, IL-6, and other cytokines were not measured and may be included in future research. Third, the link between lp (a) variability and CRP levels during follow-up was investigated but major adverse cardiovascular events (MACEs) were not. A future study focusing on the relationship between lp (a) and MACE is planned.

CONCLUSION

In this multicenter retrospective study, lp (a) variability was significantly associated with mean CRP levels during 1-year follow-up in post-PCI CAD patients. High lp (a) variability was found to be an independent risk factor for increased CRP levels following PCI treatment.

ARTICLE HIGHLIGHTS

Research background

Lipoprotein a [lp (a)] variability is gaining growing attention in cardiovascular disease.

Research motivation

The association between lp (a) variability and follow-up C-reactive protein (CRP) level in patients undergoing percutaneous coronary intervention (PCI) has not been investigated.

Research objectives

To explore the association between lp (a) variability and mean CRP levels within the 1^st year post-PCI.

Research methods

This was a multicenter retrospective study.

Research results

Lp (a) variability was positively related to the mean follow-up CRP in patients undergoing PCI.

Research conclusions

High variability could be considered an independent risk factor for increased post-PCI CRP level.

Research perspectives

Prospective studies are further needed to verify the results of this paper.