Statin Use and Venous Thromboembolism Incidence: A Case-Control Study in Patients Aged 75 Years and Older

Clinical Geriatrics

Gerontol Geriatr Res. 2023; 9(2): 1089.

Statin Use and Venous Thromboembolism Incidence: A Case-Control Study in Patients Aged 75 Years and Older

Lisa Guénard1,2; Aurélie Lafargue² ; Eric Ouattara³; Fabrice Bonnet4,5; Isabelle Bourdel-Marchasson2,6*

1Université de Bordeaux, UFR de Médecine, Bordeaux, France

2CHU de Bordeaux, Pole de gérontologie clinique, Bordeaux, France

3CHU de Bordeaux, Service d’information médicale, UCAIM-DIM, Bordeaux, France

4CHU de Bordeaux, Service de Médecine Interne et Maladies Infectieuses, hôpital Saint-André, CHU de Bordeaux, Bordeaux, France

5ISPED, INSERM U1219, Bordeaux Population Health Research Center, University of Bordeaux, F-33000, Bordeaux, France

6CNRS/Université de Bordeaux, CRMSB, UMR 5536, Bordeaux, France

*Corresponding author: Marchasson IB Centre Henri Choussat, Hôpital Xavier Arnozan, 33604 Pessac, France Email: Isabelle.bourdel-marchasson@chu-bordeaux.fr

Received: May 03, 2023 Accepted: May 27, 2023 Published: June 03, 2023

Abstract

Purpose: Although previous observational and controlled studies have reported a lower rate of venous thromboembolism in statin users, such studies were conducted in selected patients and their results may not be extended to an older multi-morbid population. This study aimed to investigate the influence of statin use on the risk of Venous Thromboembolic Events (VTE) in older patients.

Methods: We conducted a case-control study on patients aged 75 and older, hospitalized in the Medicine or Geriatric Departments from 2009 to 2019. Cases with a documented episode of VTE were included and compared with random controls matched for age, gender, year of hospitalization, and previous statin use. Patients under long-term anticoagulation therapy before admission were excluded.

Results: Of the 177 cases, 31 (17.5%) reported statin use, as well as 36 (20.3%) controls. Statin use was not significantly associated with a lower rate of VTE (Odds Ratio [OR] 0.8 95% Confidence Interval [CI] [0.5-1.4], p=0.5). Adjustments for age, sex, atherosclerotic disease, major risk factors for VTE, and use of antiplatelet agents did not alter the results. Neither aspirin (OR 1.0 95% CI 0.6-1.6], p=1.0) nor other antiplatelet therapies (OR 0.6 95% CI [0.3-1.3], p=0.2) were associated with a reduced rate of VTE. Statin users had significantly more comorbidities, such as atherosclerotic disease, hypertension, and diabetes than non-users.

Conclusion: This case-control study does not support a protective effect of statins on VTE risk in older inpatients.

Keywords: Statin; Venous thromboembolism; Older; Case-control; Comorbidities

Introduction

Venous Thrombo Embolism (VTE), including Deep Vein Thrombosis (DVT) and Pulmonary Embolism (PE) is the third leading cause of cardiovascular disease in western countries and is strongly associated with age. The incidence of VTE is 3-4/1,000 per year among those aged 65 years and older, which increases to 1% in the oldest [1,2]. Despite a well-defined therapeutic strategy based on anticoagulation [3] morbidity and mortality related to VTE remain substantial in older patients, and bleeding complications under treatment are more frequent than in younger patients [4-6]. Thus, reinforcing primary and secondary prevention of VTE is of growing interest.

Primary prevention mainly relies on low-dose-heparin and low-molecular-weight-heparin prophylaxis for at-risk situations, yet these therapies do not eliminate the VTE risk [3]. Anticoagulant therapy is effective for preventing both incident and recurrent VTE, and the guidelines recommend extended anticoagulant use after a first VTE episode in patients at high risk of recurrence [3,7,8]. Nevertheless, anticoagulant therapy is associated with bleeding complications that may limit its use in older and multi-morbid patients. Identifying older individuals with a favorable risk-benefit ratio for long-term anticoagulation therapy may be challenging [9,10]. Older adults, who are at higher risk for both VTE and bleeding complications, may benefit from a preventive treatment that would not increase the risk of bleeding.

JUPITER was a randomized controlled interventional trial that investigated the effectiveness of a high potency statin, rosuvastatin 20mg daily, in the primary prevention of vascular events, including VTE. The authors reported a significant 40% risk reduction in the occurrence of VTE in healthy subjects receiving rosuvastatin [11]. This result was consistent with other studies suggesting that the benefits of statin therapy rely not simply on its lipid-lowering effect. Statins may exert pleiotropic effects by reducing inflammatory markers, inhibiting platelet activation, improving endothelial function, and interfering with the blood coagulation system [12]. Thus, the hypothesis of a protective effect of statins in non-atherosclerotic vascular diseases, such as VTE, has been proposed.

Several observational studies have reported a significant decrease in the risk of VTE in statin-treated patients, with different relative risk reductions, ranging from 20% to 60% [13-15]. Most of these studies included young or middle-aged patients with few comorbidities. In the post-hoc analysis of the PROSPER study, the authors found a lack of a protective effect of 40 mg pravastatin daily on first VTE incidence in subjects aged 70-82 years [16]. Whether statin use is associated with a lower risk of VTE in all patient subgroups remains controversial. Furthermore, data in older patients are scarce, even though they represent a high proportion of VTE cases (70% in those aged ≥60 years) [2].

The objective of this case-control study was to evaluate the association between statin use and VTE in hospitalized patients aged 75 years or older presenting with acute VTE.

Methods

Design and settings

This monocentric cross-sectional case-control study compared hospitalized patients aged 75 years and older, presenting with acute symptomatic VTE, according to the use of statins. All patients were hospitalized between 2009 and 2019 in two wards (Internal or Geriatric Medicine).

Our primary objective was to evaluate the association between statin use and the occurrence of VTE. The primary outcome was a comparison of the presence of recent VTE according to statin use in case and control patients who were not undergoing anticoagulation therapy, estimated with odds ratios (ORs) and 95% Confidence Intervals (CIs). The secondary outcomes were to describe VTE cases (VTE location, provoked and idiopathic events, transient risk factors, comorbidities, and co-medications) and to compare statin users to non-users (comorbidities and co-medications).

This protocol was approved by the hospital ethical review board.

Study Population

A list of potential cases and their matched controls were obtained in a request to the medical information department of CHU of Bordeaux. Medical records were extracted from the hospital database using the ICD-10 classification (International Classification of Diseases). The eligibility of all patients was reviewed by the study investigators.

Cases were defined as patients aged 75 years and older diagnosed with symptomatic PE, proximal DVT of the lower limb, or both. The VTE diagnosis had to be confirmed with validated objective imaging methods. DVT was defined as the acute onset of leg pain or swelling, associated with incomplete compressibility of a proximal venous segment on ultrasonography. PE was defined using the association between the clinical symptoms (dyspnea or oxygen requirement, chest pain, hemoptysis, or syncope) and an intraluminal filling defect on contrast spiral Computed Tomography (CT), or a high probability ventilation/perfusion lung scan when CT was contraindicated.

A control subject matched to the same age, gender, and hospitalization year was randomly assigned to each case subject. Cases and their matched controls had to be hospitalized during the same year in one of the participating units. Controls were hospitalized patients who were not diagnosed with VTE.

The exclusion criteria for cases were venous thrombosis at an atypical location (cerebral, portal, jugular, or inferior or superior vena cava thrombosis), DVT of the upper limbs, isolated distal DVT of the lower limb, and superficial venous thrombosis. Controls were excluded if they had had a previous episode of objectively confirmed VTE. Cases and controls were excluded if they had been treated with anticoagulant therapy before admission.

Data Collection

Baseline data were collected from the hospital database by the study investigator, including demographic information (age and gender), comorbid conditions (history of DVT, myocardial infarction or angina pectoris, congestive heart failure, peripheral vascular disease, cerebrovascular accident or transient ischemic attack, hypertension, chronic pulmonary disease, liver disease, renal insufficiency, diabetes, cancer, and dementia). The Charlson Comorbidity Index was calculated for each patient [17].

The use of medication at the time of diagnosis was recorded: statin (drug and dose), other lipid-lowering drugs, oral antiplatelet therapy, antihypertensive therapy, oral anti-diabetic and insulin therapy, psychotropic medications, and cancer treatment, such as current chemotherapy, radiotherapy, or hormone therapy.

Biological data were recorded at the time of hospital admission, including plasma creatinine (μg/l) and C-reactive protein (mg/l) levels. Creatinine clearance was calculated using the CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration) formula. Transient risk factors for VTE were recorded and included surgery within the past 3 months, fracture of a lower limb within the past 3 months, recent immobilization (defined as =72 hours nonsurgical bedridden status in the past 15 days), cancer (active cancer, under treatment or not, or cancer in remission for less than 6 months), and severe infection [18]. Cases without any transient risk factors were considered to have idiopathic VTE.

Statin use was defined as the use of statin therapy at the time of hospital admission.

Statistical Analysis

The null hypothesis was the prevalence of VTE is equal or superior in patients who use statins compared to patients who do not. The required number of subjects to test this hypothesis unilaterally, with a ratio of one control for one case, with a statistical power of 85% and an alpha risk of 0.05, was estimated from a previous case-control study [13]. In the latest study that included 377 patients, the OR for VTE was 0.42 in statin users compared to non-users. In this study, a minimal detectable OR of 0.45 was chosen to detect a difference and a large effect size, which may be more likely to have practical significance. The prevalence of statin use was tailored according to recent observational studies [19] and the usually observed prevalence in geriatrics wards, resulting in an expected prevalence of statin use in older patients of 20%. Thus, we obtained the required 332 subjects (166 in the case group and 166 in the control group).

The characteristics of the cases and controls were compared using the chi-square test for categorical data or Fisher’s exact test for paired values, when appropriate. Continuous variables were compared with Student’s t-test for paired values. The OR for VTE was estimated according to drug use in the matched cases and controls.

A pre-planned stratified analysis was performed to explore potential confounders known to be either risk factors or protective factors for VTE, including age, sex, transient major risk factors, cardiac insufficiency, coronary heart disease, cerebrovascular disease, occlusive arterial disease of the lower limb, chronic lung disease, and exposure to antiplatelet therapy (aspirin, clopidogrel, or ticagrelor). Homogeneity of the data was assessed using the Breslow-Day test, and adjusted ORs were calculated with the Mantel-Haenszel method.

The comparisons of statin users and non-users were pre-planned and performed with the chi-square and Fisher’s exact test for categorical variables and Student’s t-test for paired continuous variables. The data analysis was conducted with SPSS© software (SPSS version 23; SPSS Inc., Chicago, IL, USA). A p-value < 0.05 was considered significant.

Results

Participants

In total, 354 patients were enrolled (177 cases and 177 controls; Figure 1). All patients had been hospitalized between 2009 and 2019 at the University Hospital in Bordeaux. The baseline characteristics of the participants are summarized in Table 1.