Abstract
To assess the association and the predictive value of plasma homocysteine (Hcy) levels with LA/LAA thrombus in non-valvular Atrial fibrillation (AF) patients with low CHA2DS2-VASc score.
Eight hundred and eighty-eight consecutive patients in non-valvular AF with CHA2DS2-VASc score of 0 and 1 were enrolled. All patients routinely underwent transthoracic echocardiography and transoesophageal echocardiography. A total of thirty-two patients had LA/LAA thrombus. Compared with patients without LA/LAA thrombus, plasma Hcy levels were significantly higher in patients with LA/LAA thrombus (16.5 ± 4.8 mmol/L vs. 13.4 ± 4.1 mmol/L, P = 0.009). In multivariate analysis, Hcy was independently associated with LA/LAA thrombus (OR 1.048, 95% CI 1.007–1.090, P = 0.022). Hcy demonstrated a significant predictive value with area under the curve of 0.722 (95% CI 0.662–0.781, P < 0.001). The optimal cut-off point for Hcy predicting LA/LAA thrombus was 13.5 mmol/L (sensitivity 67%, specificity 65%). Patients with Hcy ≥13.5 mmol/L had higher prevalence of LA/LAA thrombus compared with those with Hcy <13.5 mmol/L (6.1% vs. 2.1%, P < 0.001). Elevated Hcy significantly increased the risk of LA/LAA thrombus in patients with CHA2DS2-VASc score of 0 and 1 (OR 11.789, 95% CI 1.437–96.746, P = 0.022; OR 2.256, 95% CI 1.007–5.155, P = 0.048, respectively).
Elevated plasma Hcy increases the risk of LA/LAA thrombus in non-valvular AF patients with low CHA2DS2-VASc score, thus it should be taken into account in prediction of thromboembolism.
Plasma homocysteine (Hcy) levels were significantly higher in patients with LA/LAA thrombus than those without LA/LAA thrombus.
Baseline Hcy was independently associated with LA/LAA thrombus in non-valvular atrial fibrillation patients with low CHA2DS2-VASc score.
Elevated Hcy significantly increased the risk of LA/LAA thrombus in patients with low CHA2DS2-VASc score.
Introduction
Atrial fibrillation (AF), the most common sustained cardiac arrhythmia in clinical practice, confers a series of adverse outcomes. Thromboembolism is the worst complication to increase mortality and morbidity in AF patients. Non-valvular AF is independently associated with a four- to five-fold increased risk of ischemic stroke.1 AF-related ischaemic stroke is predominately attribute to the thrombus from left atrial (LA) cavity especially from left atrial appendage (LAA). However, limited factors are available in prediction of LA/LAA thrombus by far in AF patients. CHA2DS2-VASc scoring system is the widely used scheme for stratifying ischaemic stroke risk in patients with non-valvular AF in the guidelines.2 Patients with CHA2DS2-VASc score ≥2 were defined as the high risk and would be recommended to oral anticoagulation therapy. Patients with score of 0 and 1 were defined as the low risk, while there was a proportion of these patients to suffer from ischaemic stroke. The risk of LA/LAA thrombus in AF patients with low CHA2DS2-VASc score had been always underestimated.3 It would be of great clinical significance to identify individuals at relatively high risk of LA/LAA thrombus who had low CHA2DS2-VASc score of either 0 or 1.
Homocysteine (Hcy), an intermediate amino acid produced during the metabolism of dietary methionine, is linked with a number of vascular diseases including hypertension, coronary artery disease, stroke, etc.4 Hyperhomocysteinemia is an independent risk factor for arterial and deep venous thrombosis.5 And we have recently reported that increased plasma Hcy levels are associated with early recurrence of atrial tachyarrhythmia after catheter ablation in persistent AF patients.6 Whether Hcy conveys a risk for the formation of LA/LAA thrombus in AF with low CHA2DS2-VASc score remains unknown, so the objective of this study is to assess the association and the predictive value of plasma Hcy levels with LA/LAA thrombus in non-valvular AF patients with low CHA2DS2-VASc score of 0 and 1 .
Methods
Study population
A total of 1695 consecutive patients with non-valvular AF were enrolled from the Chinese Atrial Fibrillation Registry (CAFR) study between January 2015 and December 2015. The definition and classification of AF were according to the published guideline.2 Paroxysmal AF was defined as recurrent AF that terminates spontaneously within 7 days, and persistent AF including persistent AF and long-standing persistent AF was defined as any AF episode lasting longer than 7 days or requiring termination by cardioversion. The CHA2DS2-VASc score was calculated for each patient as follows: two points were assigned for a history of stroke or transient ischaemic attack (TIA), or age ≥75 years; and 1 point was assigned for age 65–74 years, history of hypertension, diabetes mellitus, heart failure, vascular disease, and female sex.2
Patients with secondary AF, organic valvular heart diseases, congenital heart diseases, ischaemic stroke within 6 months, deep venous thrombosis, pulmonary embolism, left ventricular thrombus, hyperthyroidism, malignant tumour, severe liver/renal dysfunction, and taking drugs that significantly affect Hcy metabolism (i.e., VitB6, VitB12, folic acid) were excluded. After admission, oral anticoagulants and antiplatelet drugs were discontinued, and all the patients received subcutaneous low-molecular weight heparin twice per day. The study complied with the Declaration of Helsinki. The protocol was approved by the institutional ethics review committee, and all patients provided written informed consent.
Blood sampling and assays
Venous blood samples were obtained from the basilic vein the morning after admission (after overnight fast). Blood samples were collected in tubes containing EDTA, and plasma was separated within 60 min. All samples were assayed in the laboratory of our hospital, and laboratory personnel were blinded to the clinical status. Laboratory data, including Hcy, creatinine, and fibrinogen levels were collected and analysed. Plasma Hcy levels were quantitatively determined by cycling enzymatic method (Beckman Coulter AU5400 automatic biochemical analyser, CA, USA). The baseline glomerular filtration rate (GFR) was evaluated using the Cockcroft–Gault formula.7
Ultrasound evaluation
All patients routinely underwent transthoracic echocardiography and transoesophageal echocardiography (TEE) after admission. A Vivid 9 cardiovascular ultrasound system (GE Healthcare, PA, USA) with a 5 MHz omniplane probe was used to acquire transthoracic echocardiogram and TEE images. The left atrial diameter (LAD), left ventricular ejection fraction (LVEF) and other parameters were measured from the parasternal M-mode or 2D images. Before the TEE study, the procedure was explained in detail, and written informed consent was obtained. Lidocaine hydrochloride spray was used for local anaesthesia, sedation, or anaesthesia was not applied. Left atrial cavity and LAA were carefully evaluated in multiple views to inspect for the thrombus. Left atrial/LAA thrombus was defined as a fixed or mobile, well-circumscribed echogenic mass with a unique echotexture contrasting with the adjacent or underlying myocardium. All measurements were performed and interpreted by experienced physicians who were blind to the study.
Statistical analysis
All statistical analysis was performed with SPSS19.0 (Chicago, IL, USA). Values were presented as mean ± SD for normally distributed continuous variables, and proportions for categorical variables. The differences between continuous values were assessed using an unpaired two-tailed t-test for normally distributed continuous variables, a Mann–Whitney test for skewed variables. Categorical variables were compared using χ2 test or Fisher’s exact test if necessary. Multivariate logistic regression analysis, which included variables with a P-value < 0.05 found on univariate analysis, was performed to identify the predictors of LA/LAA thrombus. All ORs were given with the 95% confidence interval (CI).
A receiver operating characteristic (ROC) curve was used to test the ability of Hcy to predict LA/LAA thrombus and to identify the optimal cut-off value, with the area under the curve (AUC) for determining the predictive value. The optimal cut-off value was defined as the point with the highest sum of sensitivity and specificity. All probability values were two-sided and a P-value < 0.05 was considered statistically significant.
Results
Baseline characteristics of patients
Of the 1695 patients enrolled, those with CHA2DS2-VASc score ≥2 were excluded. A total of 888 patients in non-valvular AF with CHA2DS2-VASc score 0 and 1 were therefore included in this study. The baseline characteristics of the study population are showed in Table 1. Two hundred and fourteen (24%) patients had hypertension, 56 (6%) had diabetes mellitus, 63(7%) had coronary artery disease, 10 (1.1%) had congestive heart failure defined as New York Heart Association (NYHA) class II or higher, and 120 (14%) had dyslipidaemia. Notably, none of the patients had previous history of stroke/TIA.
Baseline characteristics of patients with or without LA/LAA thrombus
| Variable | Total (n = 888) | Thrombus (n = 32) | No thrombus (n = 856) | P-value |
|---|---|---|---|---|
| Age, year | 58 ± 8 | 62 ± 9 | 58 ± 7 | 0.011 |
| Male sex, n (%) | 767 (86%) | 23 (72%) | 744 (87%) | 0.005 |
| BMI, kg/m2 | 25 ± 5 | 26 ± 6 | 25 ± 5 | 0.063 |
| Persistent AF, n (%) | 354 (40%) | 14 (44%) | 340 (40%) | 0.648 |
| AF history, months | 48.9 ± 49.6 | 50.2 ± 47.5 | 48.8 ± 51.3 | 0.259 |
| Hypertension, n (%) | 214 (24%) | 8 (25%) | 206 (24%) | 0.903 |
| Diabetes mellitus, n (%) | 56 (6%) | 2 (6%) | 54 (6%) | 0.989 |
| Coronary artery disease, n (%) | 63 (7%) | 3 (9%) | 60 (7%) | 0.609 |
| Chronic heart failure, n (%) | 10 (1.1%) | 0 (0) | 10 (1.2%) | 0.539 |
| Previous stroke/TIA, n (%) | 0 (0) | 0 (0) | 0 (0) | – |
| Dyslipidaemia, n (%) | 120 (14%) | 5 (16%) | 115 (13%) | 0.722 |
| LAD, mm | 39 ± 6 | 46 ± 7 | 39 ± 6 | <0.001 |
| LVEF, % | 63 ± 9 | 58 ± 10 | 63 ± 9 | 0.001 |
| Oral antiplatelet drug, n (%) | 143 (16%) | 6 (19%) | 137 (16%) | 0.271 |
| CHA2DS2-VASc Score | 0.5 ± 0.2 | 0.8 ± 0.3 | 0.5 ± 0.2 | 0.012 |
| Creatinine, μmol/L | 79 ± 10 | 84 ± 12 | 78 ± 10 | 0.094 |
| GFR, mL/min | 83 ± 14 | 86 ± 16 | 83 ± 14 | 0.367 |
| Fibrinogen, mg/dL | 3.2 ± 0.5 | 3.1 ± 0.6 | 3.2 ± 0.5 | 0.421 |
| Hcy, μmol/L | 13.5 ± 4.5 | 16.5 ± 4.8 | 13.4 ± 4.1 | 0.009 |
| Variable | Total (n = 888) | Thrombus (n = 32) | No thrombus (n = 856) | P-value |
|---|---|---|---|---|
| Age, year | 58 ± 8 | 62 ± 9 | 58 ± 7 | 0.011 |
| Male sex, n (%) | 767 (86%) | 23 (72%) | 744 (87%) | 0.005 |
| BMI, kg/m2 | 25 ± 5 | 26 ± 6 | 25 ± 5 | 0.063 |
| Persistent AF, n (%) | 354 (40%) | 14 (44%) | 340 (40%) | 0.648 |
| AF history, months | 48.9 ± 49.6 | 50.2 ± 47.5 | 48.8 ± 51.3 | 0.259 |
| Hypertension, n (%) | 214 (24%) | 8 (25%) | 206 (24%) | 0.903 |
| Diabetes mellitus, n (%) | 56 (6%) | 2 (6%) | 54 (6%) | 0.989 |
| Coronary artery disease, n (%) | 63 (7%) | 3 (9%) | 60 (7%) | 0.609 |
| Chronic heart failure, n (%) | 10 (1.1%) | 0 (0) | 10 (1.2%) | 0.539 |
| Previous stroke/TIA, n (%) | 0 (0) | 0 (0) | 0 (0) | – |
| Dyslipidaemia, n (%) | 120 (14%) | 5 (16%) | 115 (13%) | 0.722 |
| LAD, mm | 39 ± 6 | 46 ± 7 | 39 ± 6 | <0.001 |
| LVEF, % | 63 ± 9 | 58 ± 10 | 63 ± 9 | 0.001 |
| Oral antiplatelet drug, n (%) | 143 (16%) | 6 (19%) | 137 (16%) | 0.271 |
| CHA2DS2-VASc Score | 0.5 ± 0.2 | 0.8 ± 0.3 | 0.5 ± 0.2 | 0.012 |
| Creatinine, μmol/L | 79 ± 10 | 84 ± 12 | 78 ± 10 | 0.094 |
| GFR, mL/min | 83 ± 14 | 86 ± 16 | 83 ± 14 | 0.367 |
| Fibrinogen, mg/dL | 3.2 ± 0.5 | 3.1 ± 0.6 | 3.2 ± 0.5 | 0.421 |
| Hcy, μmol/L | 13.5 ± 4.5 | 16.5 ± 4.8 | 13.4 ± 4.1 | 0.009 |
AF, atrial fibrillation; TIA, transient ischaemic attack; LAD, left atrial diameter; LVEF, left ventricular ejection fraction; GFR, glomerular filtration rate; Hcy, homocysteine.
Baseline characteristics of patients with or without LA/LAA thrombus
| Variable | Total (n = 888) | Thrombus (n = 32) | No thrombus (n = 856) | P-value |
|---|---|---|---|---|
| Age, year | 58 ± 8 | 62 ± 9 | 58 ± 7 | 0.011 |
| Male sex, n (%) | 767 (86%) | 23 (72%) | 744 (87%) | 0.005 |
| BMI, kg/m2 | 25 ± 5 | 26 ± 6 | 25 ± 5 | 0.063 |
| Persistent AF, n (%) | 354 (40%) | 14 (44%) | 340 (40%) | 0.648 |
| AF history, months | 48.9 ± 49.6 | 50.2 ± 47.5 | 48.8 ± 51.3 | 0.259 |
| Hypertension, n (%) | 214 (24%) | 8 (25%) | 206 (24%) | 0.903 |
| Diabetes mellitus, n (%) | 56 (6%) | 2 (6%) | 54 (6%) | 0.989 |
| Coronary artery disease, n (%) | 63 (7%) | 3 (9%) | 60 (7%) | 0.609 |
| Chronic heart failure, n (%) | 10 (1.1%) | 0 (0) | 10 (1.2%) | 0.539 |
| Previous stroke/TIA, n (%) | 0 (0) | 0 (0) | 0 (0) | – |
| Dyslipidaemia, n (%) | 120 (14%) | 5 (16%) | 115 (13%) | 0.722 |
| LAD, mm | 39 ± 6 | 46 ± 7 | 39 ± 6 | <0.001 |
| LVEF, % | 63 ± 9 | 58 ± 10 | 63 ± 9 | 0.001 |
| Oral antiplatelet drug, n (%) | 143 (16%) | 6 (19%) | 137 (16%) | 0.271 |
| CHA2DS2-VASc Score | 0.5 ± 0.2 | 0.8 ± 0.3 | 0.5 ± 0.2 | 0.012 |
| Creatinine, μmol/L | 79 ± 10 | 84 ± 12 | 78 ± 10 | 0.094 |
| GFR, mL/min | 83 ± 14 | 86 ± 16 | 83 ± 14 | 0.367 |
| Fibrinogen, mg/dL | 3.2 ± 0.5 | 3.1 ± 0.6 | 3.2 ± 0.5 | 0.421 |
| Hcy, μmol/L | 13.5 ± 4.5 | 16.5 ± 4.8 | 13.4 ± 4.1 | 0.009 |
| Variable | Total (n = 888) | Thrombus (n = 32) | No thrombus (n = 856) | P-value |
|---|---|---|---|---|
| Age, year | 58 ± 8 | 62 ± 9 | 58 ± 7 | 0.011 |
| Male sex, n (%) | 767 (86%) | 23 (72%) | 744 (87%) | 0.005 |
| BMI, kg/m2 | 25 ± 5 | 26 ± 6 | 25 ± 5 | 0.063 |
| Persistent AF, n (%) | 354 (40%) | 14 (44%) | 340 (40%) | 0.648 |
| AF history, months | 48.9 ± 49.6 | 50.2 ± 47.5 | 48.8 ± 51.3 | 0.259 |
| Hypertension, n (%) | 214 (24%) | 8 (25%) | 206 (24%) | 0.903 |
| Diabetes mellitus, n (%) | 56 (6%) | 2 (6%) | 54 (6%) | 0.989 |
| Coronary artery disease, n (%) | 63 (7%) | 3 (9%) | 60 (7%) | 0.609 |
| Chronic heart failure, n (%) | 10 (1.1%) | 0 (0) | 10 (1.2%) | 0.539 |
| Previous stroke/TIA, n (%) | 0 (0) | 0 (0) | 0 (0) | – |
| Dyslipidaemia, n (%) | 120 (14%) | 5 (16%) | 115 (13%) | 0.722 |
| LAD, mm | 39 ± 6 | 46 ± 7 | 39 ± 6 | <0.001 |
| LVEF, % | 63 ± 9 | 58 ± 10 | 63 ± 9 | 0.001 |
| Oral antiplatelet drug, n (%) | 143 (16%) | 6 (19%) | 137 (16%) | 0.271 |
| CHA2DS2-VASc Score | 0.5 ± 0.2 | 0.8 ± 0.3 | 0.5 ± 0.2 | 0.012 |
| Creatinine, μmol/L | 79 ± 10 | 84 ± 12 | 78 ± 10 | 0.094 |
| GFR, mL/min | 83 ± 14 | 86 ± 16 | 83 ± 14 | 0.367 |
| Fibrinogen, mg/dL | 3.2 ± 0.5 | 3.1 ± 0.6 | 3.2 ± 0.5 | 0.421 |
| Hcy, μmol/L | 13.5 ± 4.5 | 16.5 ± 4.8 | 13.4 ± 4.1 | 0.009 |
AF, atrial fibrillation; TIA, transient ischaemic attack; LAD, left atrial diameter; LVEF, left ventricular ejection fraction; GFR, glomerular filtration rate; Hcy, homocysteine.
Assessment of left atrial/left atrial appendage thrombus
A total of thirty-two patients had LA/LAA thrombus, and only three had them in the LA cavity. The clinical characteristics of patients with and without LA/LAA thrombus are showed in Table 1. Patients with LA/LAA thrombus were more likely to be older, female sex, and had larger left atrial dimension (LAD), lower left ventricular ejection fraction (LVEF), and higher CHA2DS2-VASc score.
Hcy in prediction of left atrial/left atrial appendage thrombus
It can be seen from Table 1 and Figure 1, plasma Hcy levels were significantly higher in patients with LA/LAA thrombus than in those without LA/LAA thrombus (16.5 ± 4.8 mmol/L vs. 13.4 ± 4.1 mmol/L, P = 0.009). Based on ROC curve analysis in Figure 2, Hcy demonstrated a significant predictive value with AUC of 0.722 (95% CI 0.662–0.781, P < 0.001). And the optimal cut-off point for Hcy predicting LA/LAA thrombus was 13.5 mmol/L, with a sensitivity of 67% and a specificity of 65%.
Plasma Hcy levels in patients with LA/LAA thrombus compared with those without LA/LAA thrombus.
Receiver operating characteristic curve for Hcy in predicting LA/LAA thrombus.
Clinical and echocardiographic factors were tested to predict LA/LAA thrombus using the logistic regression univariate and multivariate analysis (Table 2). Univariate analysis demonstrated that Hcy (P = 0.016), age (P = 0.011), male sex (P = 0.019), LAD (P = 0.008), LVEF (P = 0.046), and CHA2DS2-VASc score (P = 0.015) were associated with LA/LAA thrombus. Persistent AF type and creatinine levels had no correlation with LA/LAA thrombus. In multivariate analysis, Hcy remained a significantly independent predictor of LA/LAA thrombus (OR 1.048, 95% CI 1.007–1.090, P = 0.022).
Clinical factors associated with LA/LAA thrombus
| Variable | Univariate analysis | Multivariate analysis | ||||
|---|---|---|---|---|---|---|
| OR | 95% CI | P-value | OR | 95% CI | P-value | |
| Age | 1.054 | 1.012–1.098 | 0.011 | 1.031 | 0.983–1.081 | 0.205 |
| Male sex | 0.385 | 0.174–0.853 | 0.019 | 0.578 | 0.219–1.522 | 0.267 |
| Persistent AF type | 1.180 | 0.579–2.405 | 0.648 | |||
| LAD | 1.065 | 1.017–1.116 | 0.008 | 1.100 | 1.061–1.140 | 0.000 |
| LVEF | 0.985 | 0.970–1.000 | 0.046 | 0.934 | 0.909–0.960 | 0.000 |
| CHA2DS2-VASc score | 2.732 | 1.214–6.150 | 0.015 | 1.694 | 0.683–4.200 | 0.255 |
| Creatinine | 1.009 | 0.998–1.019 | 0.114 | |||
| Hcy | 1.044 | 1.008–1.082 | 0.016 | 1.048 | 1.007–1.090 | 0.022 |
| Variable | Univariate analysis | Multivariate analysis | ||||
|---|---|---|---|---|---|---|
| OR | 95% CI | P-value | OR | 95% CI | P-value | |
| Age | 1.054 | 1.012–1.098 | 0.011 | 1.031 | 0.983–1.081 | 0.205 |
| Male sex | 0.385 | 0.174–0.853 | 0.019 | 0.578 | 0.219–1.522 | 0.267 |
| Persistent AF type | 1.180 | 0.579–2.405 | 0.648 | |||
| LAD | 1.065 | 1.017–1.116 | 0.008 | 1.100 | 1.061–1.140 | 0.000 |
| LVEF | 0.985 | 0.970–1.000 | 0.046 | 0.934 | 0.909–0.960 | 0.000 |
| CHA2DS2-VASc score | 2.732 | 1.214–6.150 | 0.015 | 1.694 | 0.683–4.200 | 0.255 |
| Creatinine | 1.009 | 0.998–1.019 | 0.114 | |||
| Hcy | 1.044 | 1.008–1.082 | 0.016 | 1.048 | 1.007–1.090 | 0.022 |
AF, atrial fibrillation; LAD, left atrial diameter; LVEF, left ventricular ejection fraction; Hcy, homocysteine; CI, confidence interval.
Clinical factors associated with LA/LAA thrombus
| Variable | Univariate analysis | Multivariate analysis | ||||
|---|---|---|---|---|---|---|
| OR | 95% CI | P-value | OR | 95% CI | P-value | |
| Age | 1.054 | 1.012–1.098 | 0.011 | 1.031 | 0.983–1.081 | 0.205 |
| Male sex | 0.385 | 0.174–0.853 | 0.019 | 0.578 | 0.219–1.522 | 0.267 |
| Persistent AF type | 1.180 | 0.579–2.405 | 0.648 | |||
| LAD | 1.065 | 1.017–1.116 | 0.008 | 1.100 | 1.061–1.140 | 0.000 |
| LVEF | 0.985 | 0.970–1.000 | 0.046 | 0.934 | 0.909–0.960 | 0.000 |
| CHA2DS2-VASc score | 2.732 | 1.214–6.150 | 0.015 | 1.694 | 0.683–4.200 | 0.255 |
| Creatinine | 1.009 | 0.998–1.019 | 0.114 | |||
| Hcy | 1.044 | 1.008–1.082 | 0.016 | 1.048 | 1.007–1.090 | 0.022 |
| Variable | Univariate analysis | Multivariate analysis | ||||
|---|---|---|---|---|---|---|
| OR | 95% CI | P-value | OR | 95% CI | P-value | |
| Age | 1.054 | 1.012–1.098 | 0.011 | 1.031 | 0.983–1.081 | 0.205 |
| Male sex | 0.385 | 0.174–0.853 | 0.019 | 0.578 | 0.219–1.522 | 0.267 |
| Persistent AF type | 1.180 | 0.579–2.405 | 0.648 | |||
| LAD | 1.065 | 1.017–1.116 | 0.008 | 1.100 | 1.061–1.140 | 0.000 |
| LVEF | 0.985 | 0.970–1.000 | 0.046 | 0.934 | 0.909–0.960 | 0.000 |
| CHA2DS2-VASc score | 2.732 | 1.214–6.150 | 0.015 | 1.694 | 0.683–4.200 | 0.255 |
| Creatinine | 1.009 | 0.998–1.019 | 0.114 | |||
| Hcy | 1.044 | 1.008–1.082 | 0.016 | 1.048 | 1.007–1.090 | 0.022 |
AF, atrial fibrillation; LAD, left atrial diameter; LVEF, left ventricular ejection fraction; Hcy, homocysteine; CI, confidence interval.
Hyperhomocysteinemia, CHA2DS2-VASc score, and LA/LAA thrombus
The prevalence of LA/LAA thrombus according to different Hcy and CHA2DS2-VASc score is summarized in Figure 3, Table 3. Patients with Hcy ≥13.5 mmol/L had higher prevalence of LA/LAA thrombus compared with those with Hcy <13.5 mmol/L (6.1% (20/326) vs. 2.1% (12/562), P < 0.001). In the patients with CHA2DS2-VASc score of 0 (n = 416), LA/LAA thrombus was detected in 4.4% (7/159) of patients whose plasma Hcy levels were ≥13.5 mmol/L, whereas in only 0.4% (1/257) of patients whose plasma Hcy levels were <13.5 mmol/L. In the patients with CHA2DS2-VASc score of 1 (n = 472), the prevalence of LA/LAA thrombus was 7.8% (13/167) in patients with Hcy ≥13.5 mmol/L, and 3.6% (11/305) in those with Hcy <13.5 mmol/L. As shown in Table 3, hyperhomocysteinemia significantly increased the risk of LA/LAA thrombus in patients with CHA2DS2-VASc score of 0 and 1 (OR 11.789, 95% CI 1.437–96.746, P = 0.022; OR 2.256, 95% CI 1.007–5.155, P = 0.048, respectively).
Hyperhomocysteinemia and risk of LA/LAA thrombus in different CHA2DS2-VASc score
| CHA2DS2-VASc score | n | Prevalence of LA/LAA thrombus (hyperhomocysteinemia vs. normal) | OR (hyperhomocysteinemia vs. normal) | 95% CI | P-value |
|---|---|---|---|---|---|
| 0 | 416 | 4.4%(7/159) vs. 0.4%(1/257) | 11.789 | 1.437–96.746 | 0.022 |
| 1 | 472 | 7.8%(13/167) vs. 3.6%(11/305) | 2.256 | 1.007–5.155 | 0.048 |
| CHA2DS2-VASc score | n | Prevalence of LA/LAA thrombus (hyperhomocysteinemia vs. normal) | OR (hyperhomocysteinemia vs. normal) | 95% CI | P-value |
|---|---|---|---|---|---|
| 0 | 416 | 4.4%(7/159) vs. 0.4%(1/257) | 11.789 | 1.437–96.746 | 0.022 |
| 1 | 472 | 7.8%(13/167) vs. 3.6%(11/305) | 2.256 | 1.007–5.155 | 0.048 |
Hyperhomocysteinemia was defined as Hcy ≥13.5μmol/L.
Hyperhomocysteinemia and risk of LA/LAA thrombus in different CHA2DS2-VASc score
| CHA2DS2-VASc score | n | Prevalence of LA/LAA thrombus (hyperhomocysteinemia vs. normal) | OR (hyperhomocysteinemia vs. normal) | 95% CI | P-value |
|---|---|---|---|---|---|
| 0 | 416 | 4.4%(7/159) vs. 0.4%(1/257) | 11.789 | 1.437–96.746 | 0.022 |
| 1 | 472 | 7.8%(13/167) vs. 3.6%(11/305) | 2.256 | 1.007–5.155 | 0.048 |
| CHA2DS2-VASc score | n | Prevalence of LA/LAA thrombus (hyperhomocysteinemia vs. normal) | OR (hyperhomocysteinemia vs. normal) | 95% CI | P-value |
|---|---|---|---|---|---|
| 0 | 416 | 4.4%(7/159) vs. 0.4%(1/257) | 11.789 | 1.437–96.746 | 0.022 |
| 1 | 472 | 7.8%(13/167) vs. 3.6%(11/305) | 2.256 | 1.007–5.155 | 0.048 |
Hyperhomocysteinemia was defined as Hcy ≥13.5μmol/L.
The prevalence of LA/LAA thrombus according to Hcy and CHA2DS2-VASc score.
Discussion
In this study, we for the first time found that elevated plasma Hcy increased the risk of LA/LAA thrombus, demonstrating a significant predictive value for the patients in non-valvular AF with low CHA2DS2-VASc score of 0 and 1.
Hcy is a metabolic intermediate of the methionine-cysteine pathway. The abnormal elevation of plasma Hcy levels is defined as hyperhomocysteinemia. Although there is controversy over the fact that whether Hcy is a cause or a consequence, a number of prospective cohort studies have found the moderate elevation of Hcy is an independent risk factor for cardiovascular diseases, ischaemic stroke, venous thromboembolic diseases, atherosclerosis, diabetes mellitus, etc.8–11 Previous studies also have reported the role of Hcy in AF patients with a history of ischaemic stroke.12–16
Friedman et al. first reported that Hcy levels were higher in patients with AF who had a stroke than in those who had not, and Hcy levels increased sharply in patients with AF after 65 years of age, suggesting elevated Hcy levels might be one of the explanations for the increased rate of stroke in elderly AF patients.12 Similar results were found by Cingozbay et al.13 in a group of 38 patients with non-valvular AF. Marcucci et al.14 studied 310 AF patients on oral anticoagulant treatment and observed elevated Hcy levels were an independent risk factor for ischaemic events in non-valvular AF, and a significant positive correlation was identified between Hcy levels and LAD. Loffredo et al.15 investigated 163 non-valvular AF patients and 40 AF patients with a history of ischaemic stroke for 2 years and found Hcy was independently associated with ischaemic stroke after adjusting for traditional cardiovascular and thromboembolic risk factors. Compared with patients in the first quartile of the total Hcy distribution, patients in the fourth quartile of the total Hcy distribution had a 2.73-fold increased probability of ischemic stroke. All these studies focused on non-valvular AF patients with a history of ischaemic stroke and found some similar results.
Ay et al.16 investigated the association between Hcy and LA thrombus in 42 patients with AF-related acute ischaemic stroke and found that Hcy levels were significantly higher in 20 patients with LA thrombus than 22 patients without thrombus, indicating the thrombogenic role of high Hcy. Ay’s study was different from those above studies, as it enrolled patients with AF and acute ischaemic stroke. In our study, we demonstrated that plasma Hcy levels were significantly higher in patients with LA/LAA thrombus than those without thrombus, and elevated plasma Hcy was independently associated with LA/LAA thrombus in multivariate analysis, which were in line with Ay’s results. However, the prevalence of LA/LAA thrombus was much higher in Ay’s study, mostly because they studied the patients with AF-related acute ischaemic stroke, we studied non-valvular AF patients without previous stroke/TIA and with low CHA2DS2-VASc score of 0 and 1.
The possible mechanisms linking high levels of plasma Hcy with LA/LAA thrombus remain unclear. Hcy is a well-documented marker for pro-oxidative stress and pro-inflammation17. Moderate hyperhomocysteinemia is considered as a prothrombotic condition that may favour the development of both arterial and venous thrombosis. A high level of Hcy can directly exert biological damage to vascular endothelial cells and tissues through generating oxidative stress, decreasing NO production and bioavailability, and inducing inflammatory response.18 Moreover, Hcy can also influence the thrombotic properties of the endothelium by reducing the affinity of antithrombin and inhibiting the expression of thrombomodulin-protein C complex.19 All these alterations will decrease endothelial resistance to thrombosis and promote a prothrombotic and hypercoagulable milieu, which fulfil the Virchow's triad for thrombogenesis in non-valvular AF.20
In the present study, we found that elevated Hcy conferred a greater risk of LA/LAA thrombus in AF patients with low CHA2DS2-VASc score of 0 and 1. Compared with CHA2DS2-VASc score of 1, it showed higher OR value and more significant difference in CHA2DS2-VASc score of 0. The difference might lie in the baseline characteristics of patients and the sample size. It seemed plausible that the less risk factor at baseline and the lower prevalence of LA/LAA thrombus would facilitate the effect of Hcy on LA/LAA thrombus in patients with CHA2DS2-VASc score of 0.
Atrial fibrillation is associated with increased but variable risk of stroke. Nowadays, the CHA2DS2-VASc scoring system is a well validated and widely used clinical risk prediction tool for ischaemic stroke in non-valvular AF, however, it provides only modest discrimination of risk for individual patients with C-statistic that range from 0.549 to 0.638.21 The clinical significance of the present study was to identify individuals at high risk of LA/LAA thrombus who had a low stroke risk at baseline, which should indicate an improved decision support in the care of these relatively high risk patients.
Limitations
Our study presents several limitations. Firstly, the results of this cross-sectional, clinical observational study were a single-centre experience. The conclusion may be more precise if it is further prospectively validated in an external large cohort of AF populations. The Hcy cut-off value determined in this study should be reassessed in different populations. And the patients in this study were referred for catheter ablation of AF, so the population might not represent the general AF populations. Secondly, other oxidative stress factors and coagulation factors were not measured in this study, whether they interacted with each other remained unknown. Finally, the detailed mechanism of elevated Hcy with LA/LAA thrombus in non-valvular AF patients was not been studied.
Conclusions
Elevated plasma Hcy increases the risk of LA/LAA thrombus in non-valvular AF patients with low CHA2DS2-VASc score, thus it should be taken into account in prediction of thromboembolism.
Acknowledgements
We thank Yanfei Ruan for the manuscript revision.
Funding
This work was supported by grants from the National Natural Science Foundation of China (No. 81670294, 81200141, 81470464, 81530016), and grants (2014BAI08B08, 2016YFC0900901) from the Ministry of Science and Technology of the People’s Republic of China.
Conflict of interest: none declared.
