Baseline inflammatory status affects the prognostic impact of statins in patients with peripheral arterial disease

Study population and endpoints

This study is a post-hoc analysis of the Toma-Code Registry, which is a Japanese prospective cohort of 2,321 consecutive patients with PAD who were treated with EVT in 34 hospitals between August 2014 and August 2016. The study protocol was approved by the Ethics Committee at Sakakibara Heart Institute (reference no. 14-023) and the committees of each participating facility. This study was registered in the University Hospital Medical Information Network Clinical Trials Registry (UMIN-CTR No. UMIN000015100). The main findings of this registry have been previously published. (15) After the exclusion of patients without information on CRP levels at the time of index EVT and those who were lost to follow-up, 1,974 patients, including 1,021 statin users and 953 statin non-users, were ultimately analyzed in the current study. Referring to the prior report, (16) the patients were divided into four categories depending on the CRP level at the time of EVT: low CRP (<0.1 mg/dL), intermediate-low CRP (0.1–0.3 mg/dL), intermediate-high CRP (0.3–1.0 mg/dL), and high CRP (>1.0 mg/dL) (Figure 1). A composite of outcomes including death, stroke, myocardial infarction, and major amputation as the primary endpoint of this study was compared between statin users and non-users in each CRP category. This study was conducted according to the principles of the Declaration of Helsinki, and written informed consent was obtained from all patients before study entry.

• Download figure
• Open in new tab

Figure 1. Study population

CRP = C-reactive protein; PAD = peripheral arterial disease.

Data collection and follow-up

Demographic, laboratory, angiographic, and procedural data were collected from each patient’s hospital chart or database by independent researchers, according to predetermined definitions, and all data were collectively managed by the study office. Follow-up data were obtained from hospital charts or by contacting the patients or their family members via telephone. Patients were followed-up at 1, 6, 12, 18, and 24 months after the index EVT. We evaluated patients’ baseline characteristics, including age; sex; body mass index (BMI); history of smoking; history of revascularization therapy and lower limb amputation; having other comorbidities, including CAD, cerebrovascular disease (CVD), chronic kidney disease (CKD); complete blood count [white blood cell (WBC) and platelets]; hemoglobin level; blood chemistry test results [albumin, creatine phosphokinase, and CRP levels and lipid profiles including total-, LDL-, and high-density lipoprotein (HDL)-cholesterol levels]; and clinical presentations of PAD (Rutherford classification and ankle brachial pressure index of the target limb) at the time of EVT. The prescription of medications including aspirin, P2Y12 inhibitors, cilostazol, oral anticoagulants, beta-blockers, angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers, calcium channel blockers, and statins at discharge was recorded. CKD was defined as an estimated glomerular filtration rate (eGFR) <60 mL/min/1.73 m² and urine albumin level >30 mg/g. The location and distribution of the target lesions of EVT were also recorded: iliac, femoropopliteal, below-the-knee (BTK), graft vessel, and unilateral/bilateral.

Statistical analysis

Values are expressed as the mean ± standard deviation, median and interquartile range, and percentiles, as appropriate. The independent Student’s t-test and non-parametric equivalent Mann– Whitney U test were used to compare statin users and non-users with respect to continuous variables. The Fisher’s exact test was used to evaluate categorical variables. The chi-square test was used to compare the Rutherford classification between statin users and non-users. A composite of outcomes including death, stroke, myocardial infarction, and major amputation was evaluated using Kaplan– Meier curves, differences among the four subgroups were assessed using the log-rank trend test, and differences between statin users and non-users were assessed using the log-rank test. Univariate Cox hazard regression analysis was performed to evaluate the association between statin use and prognosis after EVT. Variables reaching a level of significance of p <0.1 in the univariate analysis and considered clinically significant were included in the multivariable model.. Multivariable Cox hazard regression analysis was performed to exclude confounding factors and identify independent risk factors for the primary composite endpoint. To evaluate the hazard ratio (HR) of the primary endpoint based on statin use in the intermediate-high-CRP category, variables such as age, sex, BMI, smoking history, dyslipidemia, CKD, aortic stenosis, serum albumin level, P2Y12 inhibitor use, iliac lesion, BTK lesion, and stent implantation were used to adjust the multivariable model. For the high-CRP category, age; sex; BMI; hypertension; CKD; heart failure; CAD; CVD; hemoglobin, albumin, and LDL-cholesterol levels; bilateral target; BTK lesion; stent implantation; and procedural complications were used. Two-tailed P-values <0.05 were considered statistically significant. Statistical analyses were performed using R software, version 3.3.0 (R Foundation for Statistical Computing, Vienna, Austria).

Patient profiles

Baseline patient profiles are presented in Table 1. In the whole cohort, the average age was 73 years, 72% of the patients were males, and statin users were significantly older and more obese. Diverse comorbidities, such as hypertension, CKD, hemodialysis, and CAD, were significantly more prevalent among statin users. The statin user group included fewer patients with atrial fibrillation or the on-ambulant status but frequently received prior EVT. In terms of laboratory data, statin users had higher hemoglobin levels and lower WBC, platelet, and CRP levels. Overall, statin users received more intensive medical treatment, except with cilostazol and oral anticoagulants. Angiography revealed that statin users were likely to have fewer BTK lesions, be in lower Rutherford classes, and receive more stent implantations.

In each CRP category, statin users were significantly younger in the higher CRP categories, and the results were consistent with those of the entire cohort. The statin user group included more patients undergoing regular hemodialysis only in the lower CRP categories. Patients in the higher CRP categories had higher Rutherford classes, and the statin user group included more non-ambulatory patients only in the high-CRP category. Baseline CRP levels were significantly lower in statin users than that in non-users, only in the high-CRP category.

Clinical outcomes

Overall, during the observation period [median 316 (interquartile range: 177–411) days), statin users had a significantly lower event rate of the primary endpoint than non-users (log-rank: p < 0.001) (Figure 2A). While patients in the lower CRP categories showed a lower event rate (log-rank test for trend: p < 0.001) (Figure 2B).

• Download figure
• Open in new tab

Figure 2. Incidence of the primary endpoint regarding statin use and CRP levels

Kaplan–Meier curves for freedom from a composite of outcomes including death, stroke, myocardial infarction, and major limb amputation in patients who were treated with and without statins (A) and in those who were categorized into four different groups according to CRP levels (B). CRP, C-reactive protein; EVT, endovascular therapy; pts, patients; inter-low, intermediate-low; inter-high, intermediate-high.

Figure 3 shows the Kaplan–Meier curves between statin users and non-users in the four different CRP categories when comparing the primary endpoint. In the intermediate-high-CRP and high-CRP categories, statin users had significantly lower rates of the primary endpoint (log-rank, p = 0.02, p = 0.008, respectively). However, in the other lower CRP categories, there were no statistically significant differences between statin users and non-users. Univariate Cox regression analysis in the high-CRP category showed that BMI, CKD, CVD, albumin, alanine aminotransferase, and statin use were associated with the incidence of the primary endpoint (Table S1). The multivariable model for patients in the high-CRP category revealed that statin use was still independently correlated with the primary endpoint, even after the adjustment of covariates [HR: 0.64, 95% confidence interval (CI): 0.41–0.98; Table 2]. For the intermediate-high-CRP category, statin use was not associated with the primary endpoint (HR: 0.65, 95% CI: 0.35–1.21).

• Download figure
• Open in new tab

Figure 3. Comparison of the primary endpoint between statin users and non-users

Incidence of composite of outcomes including death, stroke, myocardial infarction, and major limb amputation in statin users (35) and non-users (gray) in the low-CRP (A), intermediate-low-CRP (B), intermediate-high-CRP (C), and high-CRP (D) categories. CRP, C-reactive protein; EVT, endovascular therapy; pts, patient.

The components of the combined endpoints were examined individually. Multivariable analysis showed that statin use was independently associated with the incidence of death from any cause (HR: 0.57; 95% CI: 0.40–0.82), only in the entire cohort. In any other category in any component, statin users did not show prognostic superiority compared to statin non-users (Table S2).

Limitations

This study had some limitations. First, the study patients who underwent EVT and CRP testing were selected; therefore, a selection bias may have occurred. There is no evidence in this study that statins were newly prescribed during the index hospitalization or were already prescribed before hospitalization. In addition, the discontinuation or new induction of statins during the observation period was not considered in this study. Information on the types and doses of statins administered to patients was not collected; therefore, we could not determine the kind or dose-dependent effect of statins on prognosis. No therapeutic guidelines were pre-specified in the current study, and the therapeutic regime was at the discretion of the attending doctors in each hospital; therefore, some patients were indicated to receive lipid-lowering therapy with strong statins but did not receive the appropriate guideline-directed therapy. Details regarding the etiology of increased CRP levels were not examined. High-CRP levels may represent active infections, malignancies, or chronic inflammatory diseases, including collagen diseases, and the effects of statins on these conditions are unknown. CRP, but not high-sensitivity CRP, was set as the biomarker for inflammatory status in this study because high-sensitivity CRP was not routinely examined in all patients with PAD who were scheduled to receive EVT in each hospital. No changes in lipid profiles or inflammatory status were observed after discharge. Therefore, it is difficult to conclude which factor, between the anti-inflammatory and lipid-lowering properties of statins, plays a more important role in decreasing the rate of adverse clinical outcomes. Prospective statin entry studies that include CRP-categorized patients are required to clarify the prognostic impact of the anti-inflammatory effects of statins.