Thrombosis Research 133 (2014) 1011–1015
Contents lists available at ScienceDirect
Thrombosis Research journal homepage: www.elsevier.com/locate/thromres
Regular Article
Unprovoked proximal venous thrombosis is associated with an increased risk of asymptomatic pulmonary embolism Anja Boc a,b, Nina Vene a, Monika Štalc a, Katarina Košmelj c, Alenka Mavri a,⁎ a b c
Department of Vascular Diseases, University Medical Centre Ljubljana, Slovenia Institute of Anatomy, Faculty of Medicine, University of Ljubljana, Slovenia Biotechnical Faculty, University of Ljubljana, Slovenia
a r t i c l e
i n f o
Article history: Received 2 January 2014 Received in revised form 2 February 2014 Accepted 14 February 2014 Available online 19 March 2014 Keywords: Pulmonary embolism Deep venous thrombosis Incidence Risk factors
a b s t r a c t Introduction: Pulmonary embolism (PE) is common in patients with deep venous thrombosis (DVT). The outcome of DVT with concomitant symptomatic PE is worse than the outcome of isolated DVT. The risk factors for DVT and simultaneous asymptomatic PE have not been systematically studied yet. Aim: To evaluate the frequency and risk factors for asymptomatic PE in patients with DVT. Patients/methods: In 155 consecutive patients with a first episode of DVT and no PE symptoms, a ventilationperfusion lung scan was performed. Body mass index (BMI) and waist-to-hip ratio (WHR) were calculated and concentrations of D-dimer, high-sensitivity CRP (hsCRP), tissue plasminogen activator (t-PA) and troponin were measured. Laboratory tests for thrombophilia were performed. Results: Asymptomatic PE was present in 36% of patients. No differences in gender, age, BMI and WHR were found between the patients with and without PE. PE was more common in patients with proximal DVT than in those with distal DVT (42% vs. 17%, p b 0.01), and in patients with unprovoked DVT compared to patients with provoked DVT (51% vs. 28%, p b 0.01). The risk of silent PE was the highest in patients with unprovoked proximal DVT (OR, 6.9; 95% CI, 2.3–21.0). Patients with asymptomatic PE had significantly higher values of D-dimer, hsCRP, t-PA and troponin than patients with isolated DVT. Conclusions: Asymptomatic PE affected more than one third of patients with a first DVT. Unprovoked proximal DVT is the most important risk factor for the occurrence of silent PE. © 2014 Elsevier Ltd. All rights reserved.
Introduction Venous thromboembolism, the collective term used for deep venous thrombosis (DVT) and pulmonary embolism (PE), is the most common vascular disease after myocardial infarction and ischemic stroke. The incidence of venous thromboembolism is 2 to 3 per 1000 persons per year and increases with age. Approximately one third of patients with symptomatic venous thromboembolism manifest PE, whereas two thirds manifest DVT [1]. The signs and symptoms of DVT and PE are nonspecific and some patients may have clinically silent PE [2]. A systematic review of the literature showed that an asymptomatic PE occurs in at
Abbreviations: aCL, anticardiolipin antibodies; anti-beta2 GPI, antibodies against beta2-glycoprotein I; BMI, body mass index; CT, computed tomography; DVT, deep venous thrombosis; hsCRP, high-sensitivity C-reactive protein; LA, lupus anticoagulants; OR, odds ratio; PE, pulmonary embolism; t-PA, tissue type plasminogen activator; WHR, waist-tohip ratio. ⁎ Corresponding author at: Department of Vascular Diseases, University Medical Centre Ljubljana, Zaloška 7, SI-1000 Ljubljana, Slovenia. Tel.: +386 1 522 80 80; fax: +386 1 28 33 155. E-mail address: [email protected] (A. Mavri).
http://dx.doi.org/10.1016/j.thromres.2014.02.033 0049-3848/© 2014 Elsevier Ltd. All rights reserved.
least one third of the patients with acute DVT [3]. Those patients are at higher risk of recurrent PE [4–7]. It was shown recently that the risk of symptomatic PE is increased almost 5-fold in patients with silent PE in the first 2 weeks of anticoagulant therapy [8]. Although DVT and PE are generally considered as manifestations of the same disease, it is well established that patients presenting with symptomatic PE are at higher risk of fatal recurrent PE than patients presenting with DVT alone [9]. Although routine screening for asymptomatic PE in patients with DVT is not performed in clinical practice, it seems possible that if patients with a high probability of asymptomatic PE could be identified, they might benefit from screening. So far it has been demonstrated that the prevalence of silent PE in patients with proximal DVT is higher than in those with distal DVT, and it was shown recently that the silent PE occurred frequently in patients with unprovoked DVT and in patients with coexisting heart disease [10]. In patients with symptomatic venous thromboembolism also, the role of some biomarkers in the risk stratification was studied. Increased levels of troponin, D-dimer and white blood cell count were associated with the clot burden or mortality [11–13]. However, the role of biomarkers in asymptomatic PE has not been established yet. Therefore, the purpose of this clinical study was to evaluate the
1012
A. Boc et al. / Thrombosis Research 133 (2014) 1011–1015
frequency of silent PE and to find any association between clinical and laboratory factors and occurrence of asymptomatic PE. Patients and methods Patients From September 2008 to May 2010, 155 consecutive adult patients who were admitted to our outpatient clinic with a first episode of DVT of a lower limb were included in the study. Exclusion criteria were symptoms of PE at the time of enrolment, which was assessed by a standardized questionnaire, (presence of dyspnea, cough, pleuritic pain, haemoptysis, and collapse), or a previous episode of venous thromboembolism, pregnancy or cancer. With patients who entered the study, a detailed medical history was obtained, risk factors for DVT were assessed (injury, immobilisation, surgery, puerperium, estrogen intake, acute or chronic illness, leg varicosities, family history of venous thromboembolism in 1st degree relatives), a physical examination was performed, and body mass index (BMI) and waist-to-hip ratio (WHR) were calculated. Blood sampling was performed within 24 hours after confirmation of DVT. Venous blood was collected into siliconised glass vacuum tubes, 1:10 volume of 0.11 mol/L trisodium citrate, or into vacuum tubes without anticoagulant, and centrifuged. Aliquots of plasma and serum were stored at − 70 ° C until analysed. In plasma samples, tissue type plasminogen activator (t-PA) antigen was determined by an enzyme-linked immunosorbent assay (ImulyseTMt-PA, Biopool, Sweden), D-dimer was determined by an immunoturbidimetric assay Auto-Dimer (Axis-Shield, Sweden). In serum samples, high-sensitivity C-reactive protein (hsCRP) was determined by an immunoturbidimetric assay (Vitros® Fusion 5.1, Ortho-Clinical Diagnostics, USA), troponin I was determined by three-site sandwich chemiluminescence immunoassay (ADVIA Centaur XP, Siemens, USA). Anticardiolipin antibodies (aCL) were determined by standard enzyme-linked immunosorbent assay (Medium Costar, USA), and antibodies against beta2-glycoprotein I (anti-beta2 GPI) were determined by the home modified ELISA method. If they were present in medium or high titre, they were considered as positive and repeated after 12 weeks. All patients were initially treated with subcutaneous body-weight adjusted therapeutic doses of low molecular weight heparin, followed by warfarin, aiming for an international normalized ratio (INR) of 2.0 to 3.0, for at least 3 months. Four weeks after the discontinuation of warfarin treatment, tests for thrombophilia were performed. Antithrombin and protein C activity (Berichrom AT III and Berichrom Protein C, respectively, Siemens Healthcare Diagnostics, USA), and protein S free antigen (STA-Liatest® Free Protein S, Diagnostica Stago, France) were measured on an automated coagulation analyzer (Behring Coagulation Timer, Siemens Healthcare Diagnostics, USA), according to the instructions of the manufacturers. Polymorphisms of factor V Leiden and prothrombin G20210A were searched for by the real-time polymerase chain reaction (Custom TaqMan® SNP Genotyping Assays, Applied Biosystems, USA). Lupus anticoagulants (LA) were determined with LA-Screen and LA-Confirm reagents (Life Diagnostics Inc., USA). The test was repeated after 12 weeks in patients with present LA. As all recruited patients had suffered DVT, the patients with positive aCL, anti-beta2 GPI or LA on two occasions fulfilled the diagnostic criteria of antiphospholipid syndrome [14]. All the patients were Caucasians of Slovenian nationality. They signed an informed consent form to participate in the study. The study was approved by the Medical Ethical Committee of the Slovenian Ministry of Health. Diagnostic Methods The presence of DVT was confirmed with a venous ultrasound examination on an ATL Ultrasound, HDI 5000 duplex scanner (Bothell, USA),
with a linear 10 MHz probe. Iliac, common femoral, femoral, popliteal, and calf veins were examined, with the patient in a supine position. DVT affecting the iliac, common femoral, femoral and/or popliteal vein was defined as proximal, whereas DVT confined to the veins of the calf was defined as distal. Ventilation-perfusion scintigraphy of the lungs was performed within the first 48 hours after diagnosis of DVT. At least four views (anterior, posterior, left and right posterior oblique) were obtained on the standard gamma camera (Siemens Basicam or Siemens Symbia T2, Siemens, USA). Ventilation scintigraphy was performed with ultra fine dispersion of 99mTc-labelled carbon in inert gas (Technegas®, Cyclomedica, Australia), containing 400 MBq of 99mTc radioisotope (approximately 20–40 MBq accumulated in the lung). Perfusion scintigraphy was performed with 160 MBq of 99Tc-labelled microaggregated albumin (Macrotec, Mallinckrodt, Netherland), injected intravenously. In the presence of DVT, PE was confirmed if there were 1 or more segmental perfusion defects with normal ventilation, or 2 or more large subsegmental perfusion defects with normal ventilation [15]. Lung scans were categorised as normal if there were no perfusion defects. All scans were interpreted by experienced nuclear medicine physicians. Patients with non-diagnostic scans underwent chest computed tomography (CT) angiography. CT scans were performed on 128-row multidetector CT (Dual Source CT, SOMATOM® Definition, Siemens, Germany) using the spiral technique, during a single breath-hold period. In all patients, 80 ml of iodinated contrast agent (Ultravist 370 mg/ml, Bayer Health Care, Germany) were administered into the cubital vein. PE was confirmed if one or more intraluminal filling defects were present in the pulmonary arteries. Statistical Analysis Distribution of variables was tested with the Kolmogorov-Smirnov test. Variables with normal distribution were expressed as mean and standard deviation, and variables with asymmetric distribution as median and range between 1st and 3rd quartile. Differences between the groups were tested with the Student t-test for normally distributed data, and with the Mann–Whitney U test for asymmetrically distributed data. Categorical data were compared with the χ2 test. Predictive factors for PE were assessed with logistic regression; the univariate approach was followed by the multivariate approach. Data were analysed using the SPSS statistical package (IBM SPSS Statistics, USA) and program R version 3.0.2 [16]. Results One hundred and fifty-five consecutive patients with a first episode of acute lower limb DVT were included. Baseline characteristics are
Table 1 Baseline characteristics of the study participants. Characteristic Age (years) Gender male/female (N) BMI (kg/m2) WHR: men WHR: women DVT location proximal/distal (N) Risk factors for DVT, N (%) Injury with or without immobilisation Surgery Estrogen use and puerperium Acute or chronic illness Varicose veins Family history of venous thromboembolism Non identified
55 ± 16 91/64 29.0 ± 4.9 0.99 ± 0.05 0.86 ± 0.06 119/36 35 (22.6) 18 (11.6) 27 (17.4) 19 (12.3) 15 (9.7) 21 (13.6) 53 (34.2)
Data are expressed as mean ± SD. BMI, body mass index; WHR, waist-to-hip ratio; DVT, deep venous thrombosis.
A. Boc et al. / Thrombosis Research 133 (2014) 1011–1015
shown in Table 1. DVT was located predominantly in the proximal veins. 49 (31.6 %) patients had one identifiable risk factor, 53 (34.2 %) had two or more, whilst in 53 (34.2 %) no risk factors were identified. There was no difference in frequency of risk factors between patients with proximal and distal DVT. Asymptomatic PE was present in 56 (36.1 %) patients. In 50 patients, PE was confirmed with lung ventilation-perfusion scintigraphy. In 6 patients lung scans were non-diagnostic; therefore, CT angiography was performed. All 6 CT scans confirmed PE. Asymptomatic PE affected 40.6 % of women and 33.0 % of men (p = 0.33). There was no statistical difference in age, BMI and WHR between the patients with asymptomatic PE and the patients without PE (Table 2). The incidence of asymptomatic PE was higher in patients with proximal DVT than in those with distal DVT (42 % vs. 17 %, p b 0.01) and in patients with unprovoked DVT than in those with provoked DVT (51 % vs. 28 %, p b 0.01) (Table 2). Regarding type of risk factor, asymptomatic PE was less frequent in patients with injury or operation (p = 0.01 and p = 0.02, respectively), while for other risk factors, no significant differences in frequency for PE were observed. Levels of D-dimer, hsCRP, t-PA and troponin were statistically higher in patients with asymptomatic PE than in patients without PE (Table 2). For all baseline parameters that differed significantly between the patients with asymptomatic PE and the patients without PE, the univariate logistic regression analysis was performed. A significant odds ratio (OR) for PE was found for proximal compared to distal DVT (OR 3.62; 95 % CI, 1.49–10.23, p = 0.008), in unprovoked compared to provoked DVT (OR 2.61; 95 % CI, 1.32–5.25, p = 0.006), in hsCRP above median compared to hsCRP below median (OR 1.98; 95 % CI, 1.02–3.90, p = 0.046), and in t-PA above median compared to t-PA below median (OR 2.01; 95 % CI, 1.04–3.96, p = 0.040). For D-dimer borderline significance was observed (OR 1.79; 95 % CI, 0.93–3.51, p = 0.085), and the odds ratio for troponin was not significant. The most important parameters from the univariate logistic regression analysis (the location of DVT and the presence of risk factors) were included in the first multivariate logistic regression model. The lowest probability of asymptomatic PE was observed in patients with provoked distal DVT. Compared with those patients, patients with unprovoked distal DVT were at an approximately 2-fold (OR 2.3; 95 % CI, 1.1–4.6) higher risk of asymptomatic PE and patients with provoked proximal DVT at a 3-fold (OR 3.1; 95 % CI, 1.2–8.8) higher risk of asymptomatic PE. The highest risk of asymptomatic PE (OR 6.9; 95 % CI, 2.3–21.0) was observed in patients with unprovoked proximal DVT (Fig. 1). In the second multivariate logistic regression model, all significant predictors from univariate logistic regression were included: the location of DVT, the presence of risk factors, hsCRP, D-dimer and t-PA. Due to distributions highly skewed to the right, variables were dichotomized (above/below the median). The most important predictive factor for asymptomatic PE was the presence of risk factors (p = 0.024), while
Table 2 Clinical characteristics and levels of D-dimer, high-sensitivity C-reactive protein (hsCRP), tissue type plasminogen activator (t-PA) and troponin in patients with asymptomatic pulmonary embolism (PE) and patients without PE. Variable
Age (years) BMI (kg/m2) WHR (rel.) Proximal/distal DVT (N) Unprovoked/ provoked DVT (N) D-dimer (μg/l) hsCRP (mg/l) t-PA (ng/ml) troponin
Asymptomatic PE
No PE
p-value
N = 56
N = 99
57 ± 18 29.5 ± 4.9 0.92 ± 0.15 50/6 27/29
54 ± 15 28.8 ± 4.9 0.93 ± 0.09 69/30 26/73
0.28 0.38 0.64 0.005 0.005
1415 (729 – 2965) 21.9 (6.6 – 35.5) 15.8 (10.8 – 19.4) 0.011 (0.006 – 0.020)
1114 (498 – 2607) 9.6 (3.9 – 26.0) 12.3 (9.0 – 17.3) 0.007 (0.006 – 0.014)
0.042 0.028 0.043 0.026
1013
Fig. 1. Probability of asymptomatic pulmonary embolism (PE) in patients with deep venous thrombosis (DVT).
biomarkers (hsCRP, D-dimer and t-PA) and DVT location did not retain significant predictive ability (Table 3). In the study, antiphospholipid syndrome was confirmed in 16 (10.3 %) patients (positive aCL in 7 patients, anti-beta2 GPI in 4 patients, both types of antibodies in 4 patients and LA in 1 patient). Asymptomatic PE was more frequent in patients with antiphospholipid syndrome than in patients without it (68.7 % vs. 31.3 %, p = 0.04). The presence of antiphospholipid syndrome raised the risk of asymptomatic PE 4.5-fold (95 % CI, 1.5–14.0). Laboratory tests for other thrombophilia were performed in 128 patients. 27 patients were not tested: 9 of them did not attend the planned blood sampling, and in 18 patients testing was omitted because of the prolongation of the anticoagulant treatment. Factor V Leiden was confirmed in 27 (21.1 %) patients; 5 (19 %) with asymptomatic PE and 22 (81 %) without PE. Prothrombin G20210A was present in 10 patients; 4 of them suffered asymptomatic PE. With 3 patients, the presence of both mutations was confirmed and one of them suffered asymptomatic PE. Neither of the two mutations affected the frequency of asymptomatic PE significantly. Antithrombin deficiency and protein C deficiency were present in one patient each, and in both cases asymptomatic PE was confirmed. None of our patients had protein S deficiency. Discussion This study showed significant differences in the occurrence of silent PE in patients with DVT according to thrombosis location and associated Table 3 Multivariate logistic regression model predicting asymptomatic pulmonary embolism on the basis of location of DVT, risk factors for DVT and three biochemical factors: hsCRP, D-dimer and t-PA. Variables in the model
Estimate
OR
95 % CI
p-value
Location of DVT (proximal/distal) Risk factors for DVT (unprovoked/provoked DVT) hsCRP (above/below median) D-dimer (above/below median) t-PA (above/below median)
0.79
2.21
(0.82 – 6.66)
0.131
0.85
2.33
(1.12 – 4.89)
0.024
0.54
1.72
(0.81 – 3.67)
0.156
0.19
1.21
(0.56 – 2.62)
0.620
0.41
1.50
(0.74 – 3.08)
0.261
DVT, deep venous thrombosis; hsCRP, high-sensitivity C-reactive protein; t-PA, tissue type plasminogen activator
1014
A. Boc et al. / Thrombosis Research 133 (2014) 1011–1015
risk factors: in patients with unprovoked proximal DVT the incidence of asymptomatic PE was almost 7-fold higher compared to patients with provoked distal DVT. Furthermore, we found that patients with asymptomatic PE have higher levels of some biomarkers of thrombogenesis and inflammation. Data on the frequency of asymptomatic PE in patients with confirmed DVT, which was estimated in previous studies, differ substantially, from 20 to 50 %, due to study heterogeneity [4–7,17–26]. A systematic review showed that asymptomatic PE occurred in at least one-third of patients with DVT [3]. This is in accordance with our results, which show a 36 % incidence of silent PE. In the current study, asymptomatic PE was significantly more common in patients with proximal DVT than in patients with distal DVT, confirming previous results [3,10]. Patients with proximal DVT are also more likely to have symptomatic PE than those with distal DVT [20,27]. This suggests that distal thrombi may have lower embolic potential or that their clot burden is smaller than with proximal thrombi. Interestingly, the result in our study was a much higher frequency of asymptomatic PE in patients with unprovoked DVT than in patients with provoked DVT. A similar finding was reported recently, however, the data on different risk factors frequency was not provided [10]. In this report, we showed for the first time that patients with provoked distal DVT had lower probability of the asymptomatic PE than patients with unprovoked proximal DVT (Fig. 1). Thus, the latter group is at an almost 7-fold increased risk of silent PE. Knowledge of potential mechanisms involved in clot embolisation is limited. One explanation could be that clots in the venous system of the leg in patients with isolated DVT, are more resistant to embolisation due to different clot structure. Indeed, Undas and colleagues have shown that fibrin clots obtained from PE patients were more permeable, less compact, and lysed more efficiently compared with those with isolated DVT [28]. Furthermore, clot morphology and embolic potential may differ also according to the triggering factor [20], and this might partly explain the reason for a low incidence of asymptomatic PE in our patients with injury or operation. A number of biomarkers have been evaluated in patients with venous thromboembolism. D-dimer levels have been confirmed as the marker of prognosis and extent of embolic disease [29–31]. Elevated levels of troponin were associated with a 5-fold mortality in patients with acute symptomatic PE [11]. Interestingly, it was shown that hsCRP levels are higher in patients with idiopathic compared to secondary venous thromboembolism [32]. However, association between CRP levels and future venous thromboembolic events failed to be demonstrated [33,34]. Increased levels of D-dimer, hsCRP and t-PA were found in our patients with asymptomatic PE, indicating activation of coagulation and inflammatory response, and perhaps also endothelial dysfunction. However, in multivariate logistic regression analysis, the frequency of silent PE was independently associated with the presence of unprovoked DVT only, emphasising again the importance of the DVT etiology. The frequency of asymptomatic PE in patients with thrombophilia has not been assessed yet. In our study, 68.7 % of patients with newly diagnosed antiphospholipid syndrome had asymptomatic PE. Their risk of asymptomatic PE was 4.5-times higher than in other patients. In the cohort of patients with antiphospholipid syndrome, Carvera and colleagues found DVT as the most common presenting clinical manifestation, whereas symptomatic PE was less frequent [35]. However, our results showed that the majority of the patients with antiphospholipid syndrome, clinically presenting with DVT, also have PE. This suggests that PE, especially silent, may be an important clinical manifestation of antiphospholipid syndrome reflecting the systemic prothrombotic state in those patients. Screening for the presence of antiphospholipid syndrome may be useful. Factor V Leiden and prothrombin G20210A mutations did not affect the frequency of asymptomatic PE significantly. Furthermore, Factor V Leiden seemed to be less frequent in patients with asymptomatic PE. This is in concordance with the results of Rossi
and colleagues who studied the influence of thrombophilia on symptomatic PE [36]. The importance of natural coagulation inhibitor deficiency could not be assessed in our study due to the small number of such patients. Asymptomatic PE should not be underestimated. The risk of recurrent PE is increased in patients with silent PE [4–7], and Tzoran and colleagues have shown that 40 % of recurrent PE events were fatal [8]. Therefore, the balance of benefit and risk of long-term anticoagulation might be different in patients with PE or DVT and there has been some debate over the benefit of routine screening for silent PE in patients with DVT. Our study showed the highest incidence of silent PE in patients with unprovoked proximal DVT. Whether screening for PE in this subgroup of patients might be cost-beneficial should be established in prospective studies. Our study has some potential limitations. The patients received at least one therapeutic dosage of low molecular weight heparin before the blood was collected for determination of the biomarkers and this might have some influence on the results. The number of patients included was relatively low for assessing the association between the biomarkers and silent PE in the multivariate statistical models. Furthermore, the small number of the patients with confirmed thrombophilia also limits a firm conclusion on association between thrombophilia and silent PE. A larger number of patients might influence the results of multivariate analysis, and potentially elucidate the importance of thrombophilia, including antiphospholipid syndrome in silent PE. Ventilation-perfusion scintigraphy was used for PE diagnosis in this study, and therefore some patients may have had a misclassification of PE diagnosis due to limitations in the diagnostic accuracy of this method. However, a similar rate of silent PE was observed in a study by Li and colleagues, where CT angiography was used to confirm PE [10]. In conclusion, we demonstrate that patients with unprovoked proximal DVT have a high risk of asymptomatic PE. Whether aggressive and prolonged anticoagulant therapy has a more favourable outcome in patients with detected silent PE should be assessed in prospective clinical trials. Acknowledgments This study was supported by the Slovenian Ministry of Science and Technology, grant No. P3-0308. References [1] White RH. The epidemiology of venous thromboembolism. Circulation 2003;107: I4–8. [2] Kearon C. Natural history of venous thromboembolism. Circulation 2003;107:22–30. [3] Stein PD, Matta F, Musani MH, Diaczok B. Silent pulmonary embolism in patients with deep venous thrombosis: a systematic review. Am J Med 2010;123:426–31. [4] Cuppini S, Cattelan AM, Casara D, Prandoni P. Occult pulmonary embolism in patients with proximal deep venous thrombosis. Ann Ital Med Int 1991;6:1–5. [5] Meignan M, Rosso J, Gauthier H, Brunengo F, Claudel S, Sagnard L. d'Azemar P, Simonneau G, Charbonnier B. Systematic lung scans reveal a high frequency of silent pulmonary embolism in patients with proximal deep venous thrombosis. Arch Intern Med 2000;160:159–64. [6] Monreal M, Büller H, Lensing AW, Bonet M, Roncales J, Muchart J, et al. Should patients with deep vein thrombosis alone be treated as those with concomitant asymptomatic pulmonary embolism? A prospective study. Thromb Haemost 2002;88:938–42. [7] Jünger M, Diehm C, Störiko H, Hach-Wunderle V, Heidrich H, Karasch T, et al. Mobilization versus immobilization in the treatment of acute proximal deep venous thrombosis: a prospective, randomized, open, multicentre trial. Curr Med Res Opin 2006;22:593–602. [8] Tzoran I, Saharov G, Brenner B, Delsart D, Román P, Visoná A, et al. Monreal M; RIETE Investigators. Silent pulmonary embolism in patients with proximal deep vein thrombosis in the lower limbs. J Thromb Haemost 2012;10:564–71. [9] Douketis JD, Kearon C, Bates S, Duku EK, Ginsberg JS. Risk of fatal pulmonary embolism in patients with treated venous thromboembolism. JAMA 1998;279:458–62. [10] Li F, Wang X, Huang W, Ren W, Cheng J, Zhang M, et al. Risk factors associated with the occurrence of silent pulmonary embolism in patients with deep venous thrombosis of the lower limb. Phlebology 2013 [Epub 2013 May 9]. [11] Becattini C, Vedovati MC, Agnelli G. Prognostic value of troponins in acute pulmonary embolism: a meta-analysis. Circulation 2007;116:427–33. [12] Aujesky D, Roy PM, Guy M, Cornuz J, Sanchez O, Perrier A. Prognostic value of D-dimer in patients with pulmonary embolism. Thromb Haemost 2006;96:478–82.
A. Boc et al. / Thrombosis Research 133 (2014) 1011–1015 [13] Venetz C, Labarère J, Jiménez D, Aujesky D. White blood cell count and mortality in patients with acute pulmonary embolism. Am J Hematol 2013;88:677–81. [14] Miyakis S, Lockshin MD, Atsumi T, Branch DW, Brey RL, Cervera R, et al. International consensus statement on an update of the classification criteria for definite antiphospholipid syndrome (APS). J Thromb Haemost 2006;4:295–306. [15] Bajc M, Neilly JB, Miniati M, Schuemichen C, Meignan M, Jonson B. EANM guidelines for ventilation/perfusion scintigraphy: Part 2. Algorithms and clinical considerations for diagnosis of pulmonary emboli with V/P (SPECT) and MDCT. Eur J Nucl Med Mol Imaging 2009;36:1528–38. [16] R Core Team. R: A language and environment for statistical computing. Vienna: R Foundation for Statistical Computing. http://www.R-project.org; 2012 . [accessed 1 August 2013]. [17] Dorfman GS, Cronan JJ, Tupper TB, Messersmith RN, Denny DF, Lee CH. Occult pulmonary embolism: a common occurrence in deep venous thrombosis. Am J Roentgenol 1987;148:263–6. [18] Huisman MV, Büller HR, ten Cate JW, van Royen EA, Vreeken J, Kersten MJ, et al. Unexpected high prevalence of silent pulmonary embolism in patients with deep venous thrombosis. Chest 1989;95:498–502. [19] Monreal M, Rey-Joly Barroso C, Ruiz Manzano J, Salvador Tarrason R, Lafoz Naval E, Viver Manresa E. Asymptomatic pulmonary embolism in patients with deep vein thrombosis. Is it useful to take a lung scan to rule out this condition? J Cardiovasc Surg 1989;30:104–7. [20] Monreal M, Ruíz J, Olazabal A, Arias A, Roca J. Arias A, Roca J. Deep venous thrombosis and the risk of pulmonary embolism. A systematic study. Chest 1992;102:677–81. [21] Moser KM, Fedullo PF, LitteJohn JK, Crawford R. Frequent asymptomatic pulmonary embolism in patients with deep venous thrombosis. JAMA 1994;271:223–5. [22] Nielsen HK, Husted SE, Krusell LR, Fasting H, Charles P, Hansen HH. Silent pulmonary embolism in patients with deep venous thrombosis. Incidence and fate in a randomized, controlled trial of anticoagulation versus no anticoagulation. J Intern Med 1994;235:457–61. [23] Lusiani L, Visonà A, Bonanome A, Pesavento R, Zanco P. The characteristics of the thrombi of the lower limbs, as detected by ultrasonic scanning, do not predict pulmonary embolism. Chest 1996;110:996–1000. [24] López-Beret P, Pinto JM, Romero A, Orgaz A, Fontcuberta J, Oblas M. Systematic study of occult pulmonary thromboembolism in patients with deep venous thrombosis. J Vasc Surg 2001;33:515–21.
1015
[25] Monreal M, Ruiz J, Fraile M, Bonet M, Davant E, Muchart J, et al. Prospective study on the usefulness of lung scan in patients with deep vein thrombosis of the lower limbs. Thromb Haemost 2001;85:771–4. [26] Jiménez D, Díaz G, Marín E, Vidal R, Sueiro A, Yusen RD. The risk of recurrent venous thromboembolism in patients with unprovoked symptomatic deep vein thrombosis and asymptomatic pulmonary embolism. Thromb Haemost 2006;95:562–6. [27] Moser KM, LeMoine JR. Is embolic risk conditioned by location of deep venous thrombosis? Ann Intern Med 1981;94:439–44. [28] Undas A, Zawilska K, Ciesla-Dul M, Lehmann-Kopydłowska A, Skubiszak A, Ciepłuch K, et al. Altered fibrin clot structure/function in patients with idiopathic venous thromboembolism and in their relatives. Blood 2009;114:4272–8. [29] Jeebun V, Doe SJ, Singh L, Worthy SA, Forrest IA. Are clinical parameters and biomarkers predictive of severity of acute pulmonary emboli on CTPA? QJM 2010;103:91–7. [30] Grau E, Tenías JM, Soto MJ, Gutierrez MR, Lecumberri R, Pérez JL. Tiberio G; RIETE Investigators. D-dimer levels correlate with mortality in patients with acute pulmonary embolism: Findings from the RIETE registry. Crit Care Med 2007; 35:1937–41. [31] Becattini C, Lignani A, Masotti L, Forte MB, Agnelli G. D-dimer for risk stratification in patients with acute pulmonary embolism. J Thromb Thrombolysis 2012; 33:48–57. [32] Luxembourg B, Schmitt J, Humpich M, Glowatzki M, Dressler D, Seifried E, et al. Cardiovascular risk factors in idiopathic compared to risk-associated venous thromboembolism: A focus on fibrinogen, factor VIII, and high-sensitivity C-reactive protein (hs-CRP). Thromb Haemost 2009;102:668–75. [33] Hald EM, Brækkan SK, Mathiesen EB, Njølstad I, Wilsgaard T, Brox J, et al. Highsensitivity C-reactive protein is not a risk factor for venous thromboembolism: the Tromso study. Haematologica 2011;96:1189–94. [34] Fox EA, Kahn SR. The relationship between inflammation and venous thrombosis. A systematic review of clinical studies. Thromb Haemost 2005;94:362–5. [35] Cervera R, Piette JC, Font J, Khamashta MA, Shoenfeld Y, Camps MT, et al. EuroPhospholipid Project Group. Antiphospholipid syndrome: clinical and immunologic manifestations and patterns of disease expression in a cohort of 1,000 patients. Arthritis Rheum 2002;46:1019–27. [36] Rossi E, Za T, Ciminello A, Leone G, De Stefano V. The risk of symptomatic pulmonary embolism due to proximal deep venous thrombosis differs in patients with different types of inherited thrombophilia. Thromb Haemost 2008;99:1030–4.
Contents lists available at ScienceDirect
Thrombosis Research journal homepage: www.elsevier.com/locate/thromres
Regular Article
Unprovoked proximal venous thrombosis is associated with an increased risk of asymptomatic pulmonary embolism Anja Boc a,b, Nina Vene a, Monika Štalc a, Katarina Košmelj c, Alenka Mavri a,⁎ a b c
Department of Vascular Diseases, University Medical Centre Ljubljana, Slovenia Institute of Anatomy, Faculty of Medicine, University of Ljubljana, Slovenia Biotechnical Faculty, University of Ljubljana, Slovenia
a r t i c l e
i n f o
Article history: Received 2 January 2014 Received in revised form 2 February 2014 Accepted 14 February 2014 Available online 19 March 2014 Keywords: Pulmonary embolism Deep venous thrombosis Incidence Risk factors
a b s t r a c t Introduction: Pulmonary embolism (PE) is common in patients with deep venous thrombosis (DVT). The outcome of DVT with concomitant symptomatic PE is worse than the outcome of isolated DVT. The risk factors for DVT and simultaneous asymptomatic PE have not been systematically studied yet. Aim: To evaluate the frequency and risk factors for asymptomatic PE in patients with DVT. Patients/methods: In 155 consecutive patients with a first episode of DVT and no PE symptoms, a ventilationperfusion lung scan was performed. Body mass index (BMI) and waist-to-hip ratio (WHR) were calculated and concentrations of D-dimer, high-sensitivity CRP (hsCRP), tissue plasminogen activator (t-PA) and troponin were measured. Laboratory tests for thrombophilia were performed. Results: Asymptomatic PE was present in 36% of patients. No differences in gender, age, BMI and WHR were found between the patients with and without PE. PE was more common in patients with proximal DVT than in those with distal DVT (42% vs. 17%, p b 0.01), and in patients with unprovoked DVT compared to patients with provoked DVT (51% vs. 28%, p b 0.01). The risk of silent PE was the highest in patients with unprovoked proximal DVT (OR, 6.9; 95% CI, 2.3–21.0). Patients with asymptomatic PE had significantly higher values of D-dimer, hsCRP, t-PA and troponin than patients with isolated DVT. Conclusions: Asymptomatic PE affected more than one third of patients with a first DVT. Unprovoked proximal DVT is the most important risk factor for the occurrence of silent PE. © 2014 Elsevier Ltd. All rights reserved.
Introduction Venous thromboembolism, the collective term used for deep venous thrombosis (DVT) and pulmonary embolism (PE), is the most common vascular disease after myocardial infarction and ischemic stroke. The incidence of venous thromboembolism is 2 to 3 per 1000 persons per year and increases with age. Approximately one third of patients with symptomatic venous thromboembolism manifest PE, whereas two thirds manifest DVT [1]. The signs and symptoms of DVT and PE are nonspecific and some patients may have clinically silent PE [2]. A systematic review of the literature showed that an asymptomatic PE occurs in at
Abbreviations: aCL, anticardiolipin antibodies; anti-beta2 GPI, antibodies against beta2-glycoprotein I; BMI, body mass index; CT, computed tomography; DVT, deep venous thrombosis; hsCRP, high-sensitivity C-reactive protein; LA, lupus anticoagulants; OR, odds ratio; PE, pulmonary embolism; t-PA, tissue type plasminogen activator; WHR, waist-tohip ratio. ⁎ Corresponding author at: Department of Vascular Diseases, University Medical Centre Ljubljana, Zaloška 7, SI-1000 Ljubljana, Slovenia. Tel.: +386 1 522 80 80; fax: +386 1 28 33 155. E-mail address: [email protected] (A. Mavri).
http://dx.doi.org/10.1016/j.thromres.2014.02.033 0049-3848/© 2014 Elsevier Ltd. All rights reserved.
least one third of the patients with acute DVT [3]. Those patients are at higher risk of recurrent PE [4–7]. It was shown recently that the risk of symptomatic PE is increased almost 5-fold in patients with silent PE in the first 2 weeks of anticoagulant therapy [8]. Although DVT and PE are generally considered as manifestations of the same disease, it is well established that patients presenting with symptomatic PE are at higher risk of fatal recurrent PE than patients presenting with DVT alone [9]. Although routine screening for asymptomatic PE in patients with DVT is not performed in clinical practice, it seems possible that if patients with a high probability of asymptomatic PE could be identified, they might benefit from screening. So far it has been demonstrated that the prevalence of silent PE in patients with proximal DVT is higher than in those with distal DVT, and it was shown recently that the silent PE occurred frequently in patients with unprovoked DVT and in patients with coexisting heart disease [10]. In patients with symptomatic venous thromboembolism also, the role of some biomarkers in the risk stratification was studied. Increased levels of troponin, D-dimer and white blood cell count were associated with the clot burden or mortality [11–13]. However, the role of biomarkers in asymptomatic PE has not been established yet. Therefore, the purpose of this clinical study was to evaluate the
1012
A. Boc et al. / Thrombosis Research 133 (2014) 1011–1015
frequency of silent PE and to find any association between clinical and laboratory factors and occurrence of asymptomatic PE. Patients and methods Patients From September 2008 to May 2010, 155 consecutive adult patients who were admitted to our outpatient clinic with a first episode of DVT of a lower limb were included in the study. Exclusion criteria were symptoms of PE at the time of enrolment, which was assessed by a standardized questionnaire, (presence of dyspnea, cough, pleuritic pain, haemoptysis, and collapse), or a previous episode of venous thromboembolism, pregnancy or cancer. With patients who entered the study, a detailed medical history was obtained, risk factors for DVT were assessed (injury, immobilisation, surgery, puerperium, estrogen intake, acute or chronic illness, leg varicosities, family history of venous thromboembolism in 1st degree relatives), a physical examination was performed, and body mass index (BMI) and waist-to-hip ratio (WHR) were calculated. Blood sampling was performed within 24 hours after confirmation of DVT. Venous blood was collected into siliconised glass vacuum tubes, 1:10 volume of 0.11 mol/L trisodium citrate, or into vacuum tubes without anticoagulant, and centrifuged. Aliquots of plasma and serum were stored at − 70 ° C until analysed. In plasma samples, tissue type plasminogen activator (t-PA) antigen was determined by an enzyme-linked immunosorbent assay (ImulyseTMt-PA, Biopool, Sweden), D-dimer was determined by an immunoturbidimetric assay Auto-Dimer (Axis-Shield, Sweden). In serum samples, high-sensitivity C-reactive protein (hsCRP) was determined by an immunoturbidimetric assay (Vitros® Fusion 5.1, Ortho-Clinical Diagnostics, USA), troponin I was determined by three-site sandwich chemiluminescence immunoassay (ADVIA Centaur XP, Siemens, USA). Anticardiolipin antibodies (aCL) were determined by standard enzyme-linked immunosorbent assay (Medium Costar, USA), and antibodies against beta2-glycoprotein I (anti-beta2 GPI) were determined by the home modified ELISA method. If they were present in medium or high titre, they were considered as positive and repeated after 12 weeks. All patients were initially treated with subcutaneous body-weight adjusted therapeutic doses of low molecular weight heparin, followed by warfarin, aiming for an international normalized ratio (INR) of 2.0 to 3.0, for at least 3 months. Four weeks after the discontinuation of warfarin treatment, tests for thrombophilia were performed. Antithrombin and protein C activity (Berichrom AT III and Berichrom Protein C, respectively, Siemens Healthcare Diagnostics, USA), and protein S free antigen (STA-Liatest® Free Protein S, Diagnostica Stago, France) were measured on an automated coagulation analyzer (Behring Coagulation Timer, Siemens Healthcare Diagnostics, USA), according to the instructions of the manufacturers. Polymorphisms of factor V Leiden and prothrombin G20210A were searched for by the real-time polymerase chain reaction (Custom TaqMan® SNP Genotyping Assays, Applied Biosystems, USA). Lupus anticoagulants (LA) were determined with LA-Screen and LA-Confirm reagents (Life Diagnostics Inc., USA). The test was repeated after 12 weeks in patients with present LA. As all recruited patients had suffered DVT, the patients with positive aCL, anti-beta2 GPI or LA on two occasions fulfilled the diagnostic criteria of antiphospholipid syndrome [14]. All the patients were Caucasians of Slovenian nationality. They signed an informed consent form to participate in the study. The study was approved by the Medical Ethical Committee of the Slovenian Ministry of Health. Diagnostic Methods The presence of DVT was confirmed with a venous ultrasound examination on an ATL Ultrasound, HDI 5000 duplex scanner (Bothell, USA),
with a linear 10 MHz probe. Iliac, common femoral, femoral, popliteal, and calf veins were examined, with the patient in a supine position. DVT affecting the iliac, common femoral, femoral and/or popliteal vein was defined as proximal, whereas DVT confined to the veins of the calf was defined as distal. Ventilation-perfusion scintigraphy of the lungs was performed within the first 48 hours after diagnosis of DVT. At least four views (anterior, posterior, left and right posterior oblique) were obtained on the standard gamma camera (Siemens Basicam or Siemens Symbia T2, Siemens, USA). Ventilation scintigraphy was performed with ultra fine dispersion of 99mTc-labelled carbon in inert gas (Technegas®, Cyclomedica, Australia), containing 400 MBq of 99mTc radioisotope (approximately 20–40 MBq accumulated in the lung). Perfusion scintigraphy was performed with 160 MBq of 99Tc-labelled microaggregated albumin (Macrotec, Mallinckrodt, Netherland), injected intravenously. In the presence of DVT, PE was confirmed if there were 1 or more segmental perfusion defects with normal ventilation, or 2 or more large subsegmental perfusion defects with normal ventilation [15]. Lung scans were categorised as normal if there were no perfusion defects. All scans were interpreted by experienced nuclear medicine physicians. Patients with non-diagnostic scans underwent chest computed tomography (CT) angiography. CT scans were performed on 128-row multidetector CT (Dual Source CT, SOMATOM® Definition, Siemens, Germany) using the spiral technique, during a single breath-hold period. In all patients, 80 ml of iodinated contrast agent (Ultravist 370 mg/ml, Bayer Health Care, Germany) were administered into the cubital vein. PE was confirmed if one or more intraluminal filling defects were present in the pulmonary arteries. Statistical Analysis Distribution of variables was tested with the Kolmogorov-Smirnov test. Variables with normal distribution were expressed as mean and standard deviation, and variables with asymmetric distribution as median and range between 1st and 3rd quartile. Differences between the groups were tested with the Student t-test for normally distributed data, and with the Mann–Whitney U test for asymmetrically distributed data. Categorical data were compared with the χ2 test. Predictive factors for PE were assessed with logistic regression; the univariate approach was followed by the multivariate approach. Data were analysed using the SPSS statistical package (IBM SPSS Statistics, USA) and program R version 3.0.2 [16]. Results One hundred and fifty-five consecutive patients with a first episode of acute lower limb DVT were included. Baseline characteristics are
Table 1 Baseline characteristics of the study participants. Characteristic Age (years) Gender male/female (N) BMI (kg/m2) WHR: men WHR: women DVT location proximal/distal (N) Risk factors for DVT, N (%) Injury with or without immobilisation Surgery Estrogen use and puerperium Acute or chronic illness Varicose veins Family history of venous thromboembolism Non identified
55 ± 16 91/64 29.0 ± 4.9 0.99 ± 0.05 0.86 ± 0.06 119/36 35 (22.6) 18 (11.6) 27 (17.4) 19 (12.3) 15 (9.7) 21 (13.6) 53 (34.2)
Data are expressed as mean ± SD. BMI, body mass index; WHR, waist-to-hip ratio; DVT, deep venous thrombosis.
A. Boc et al. / Thrombosis Research 133 (2014) 1011–1015
shown in Table 1. DVT was located predominantly in the proximal veins. 49 (31.6 %) patients had one identifiable risk factor, 53 (34.2 %) had two or more, whilst in 53 (34.2 %) no risk factors were identified. There was no difference in frequency of risk factors between patients with proximal and distal DVT. Asymptomatic PE was present in 56 (36.1 %) patients. In 50 patients, PE was confirmed with lung ventilation-perfusion scintigraphy. In 6 patients lung scans were non-diagnostic; therefore, CT angiography was performed. All 6 CT scans confirmed PE. Asymptomatic PE affected 40.6 % of women and 33.0 % of men (p = 0.33). There was no statistical difference in age, BMI and WHR between the patients with asymptomatic PE and the patients without PE (Table 2). The incidence of asymptomatic PE was higher in patients with proximal DVT than in those with distal DVT (42 % vs. 17 %, p b 0.01) and in patients with unprovoked DVT than in those with provoked DVT (51 % vs. 28 %, p b 0.01) (Table 2). Regarding type of risk factor, asymptomatic PE was less frequent in patients with injury or operation (p = 0.01 and p = 0.02, respectively), while for other risk factors, no significant differences in frequency for PE were observed. Levels of D-dimer, hsCRP, t-PA and troponin were statistically higher in patients with asymptomatic PE than in patients without PE (Table 2). For all baseline parameters that differed significantly between the patients with asymptomatic PE and the patients without PE, the univariate logistic regression analysis was performed. A significant odds ratio (OR) for PE was found for proximal compared to distal DVT (OR 3.62; 95 % CI, 1.49–10.23, p = 0.008), in unprovoked compared to provoked DVT (OR 2.61; 95 % CI, 1.32–5.25, p = 0.006), in hsCRP above median compared to hsCRP below median (OR 1.98; 95 % CI, 1.02–3.90, p = 0.046), and in t-PA above median compared to t-PA below median (OR 2.01; 95 % CI, 1.04–3.96, p = 0.040). For D-dimer borderline significance was observed (OR 1.79; 95 % CI, 0.93–3.51, p = 0.085), and the odds ratio for troponin was not significant. The most important parameters from the univariate logistic regression analysis (the location of DVT and the presence of risk factors) were included in the first multivariate logistic regression model. The lowest probability of asymptomatic PE was observed in patients with provoked distal DVT. Compared with those patients, patients with unprovoked distal DVT were at an approximately 2-fold (OR 2.3; 95 % CI, 1.1–4.6) higher risk of asymptomatic PE and patients with provoked proximal DVT at a 3-fold (OR 3.1; 95 % CI, 1.2–8.8) higher risk of asymptomatic PE. The highest risk of asymptomatic PE (OR 6.9; 95 % CI, 2.3–21.0) was observed in patients with unprovoked proximal DVT (Fig. 1). In the second multivariate logistic regression model, all significant predictors from univariate logistic regression were included: the location of DVT, the presence of risk factors, hsCRP, D-dimer and t-PA. Due to distributions highly skewed to the right, variables were dichotomized (above/below the median). The most important predictive factor for asymptomatic PE was the presence of risk factors (p = 0.024), while
Table 2 Clinical characteristics and levels of D-dimer, high-sensitivity C-reactive protein (hsCRP), tissue type plasminogen activator (t-PA) and troponin in patients with asymptomatic pulmonary embolism (PE) and patients without PE. Variable
Age (years) BMI (kg/m2) WHR (rel.) Proximal/distal DVT (N) Unprovoked/ provoked DVT (N) D-dimer (μg/l) hsCRP (mg/l) t-PA (ng/ml) troponin
Asymptomatic PE
No PE
p-value
N = 56
N = 99
57 ± 18 29.5 ± 4.9 0.92 ± 0.15 50/6 27/29
54 ± 15 28.8 ± 4.9 0.93 ± 0.09 69/30 26/73
0.28 0.38 0.64 0.005 0.005
1415 (729 – 2965) 21.9 (6.6 – 35.5) 15.8 (10.8 – 19.4) 0.011 (0.006 – 0.020)
1114 (498 – 2607) 9.6 (3.9 – 26.0) 12.3 (9.0 – 17.3) 0.007 (0.006 – 0.014)
0.042 0.028 0.043 0.026
1013
Fig. 1. Probability of asymptomatic pulmonary embolism (PE) in patients with deep venous thrombosis (DVT).
biomarkers (hsCRP, D-dimer and t-PA) and DVT location did not retain significant predictive ability (Table 3). In the study, antiphospholipid syndrome was confirmed in 16 (10.3 %) patients (positive aCL in 7 patients, anti-beta2 GPI in 4 patients, both types of antibodies in 4 patients and LA in 1 patient). Asymptomatic PE was more frequent in patients with antiphospholipid syndrome than in patients without it (68.7 % vs. 31.3 %, p = 0.04). The presence of antiphospholipid syndrome raised the risk of asymptomatic PE 4.5-fold (95 % CI, 1.5–14.0). Laboratory tests for other thrombophilia were performed in 128 patients. 27 patients were not tested: 9 of them did not attend the planned blood sampling, and in 18 patients testing was omitted because of the prolongation of the anticoagulant treatment. Factor V Leiden was confirmed in 27 (21.1 %) patients; 5 (19 %) with asymptomatic PE and 22 (81 %) without PE. Prothrombin G20210A was present in 10 patients; 4 of them suffered asymptomatic PE. With 3 patients, the presence of both mutations was confirmed and one of them suffered asymptomatic PE. Neither of the two mutations affected the frequency of asymptomatic PE significantly. Antithrombin deficiency and protein C deficiency were present in one patient each, and in both cases asymptomatic PE was confirmed. None of our patients had protein S deficiency. Discussion This study showed significant differences in the occurrence of silent PE in patients with DVT according to thrombosis location and associated Table 3 Multivariate logistic regression model predicting asymptomatic pulmonary embolism on the basis of location of DVT, risk factors for DVT and three biochemical factors: hsCRP, D-dimer and t-PA. Variables in the model
Estimate
OR
95 % CI
p-value
Location of DVT (proximal/distal) Risk factors for DVT (unprovoked/provoked DVT) hsCRP (above/below median) D-dimer (above/below median) t-PA (above/below median)
0.79
2.21
(0.82 – 6.66)
0.131
0.85
2.33
(1.12 – 4.89)
0.024
0.54
1.72
(0.81 – 3.67)
0.156
0.19
1.21
(0.56 – 2.62)
0.620
0.41
1.50
(0.74 – 3.08)
0.261
DVT, deep venous thrombosis; hsCRP, high-sensitivity C-reactive protein; t-PA, tissue type plasminogen activator
1014
A. Boc et al. / Thrombosis Research 133 (2014) 1011–1015
risk factors: in patients with unprovoked proximal DVT the incidence of asymptomatic PE was almost 7-fold higher compared to patients with provoked distal DVT. Furthermore, we found that patients with asymptomatic PE have higher levels of some biomarkers of thrombogenesis and inflammation. Data on the frequency of asymptomatic PE in patients with confirmed DVT, which was estimated in previous studies, differ substantially, from 20 to 50 %, due to study heterogeneity [4–7,17–26]. A systematic review showed that asymptomatic PE occurred in at least one-third of patients with DVT [3]. This is in accordance with our results, which show a 36 % incidence of silent PE. In the current study, asymptomatic PE was significantly more common in patients with proximal DVT than in patients with distal DVT, confirming previous results [3,10]. Patients with proximal DVT are also more likely to have symptomatic PE than those with distal DVT [20,27]. This suggests that distal thrombi may have lower embolic potential or that their clot burden is smaller than with proximal thrombi. Interestingly, the result in our study was a much higher frequency of asymptomatic PE in patients with unprovoked DVT than in patients with provoked DVT. A similar finding was reported recently, however, the data on different risk factors frequency was not provided [10]. In this report, we showed for the first time that patients with provoked distal DVT had lower probability of the asymptomatic PE than patients with unprovoked proximal DVT (Fig. 1). Thus, the latter group is at an almost 7-fold increased risk of silent PE. Knowledge of potential mechanisms involved in clot embolisation is limited. One explanation could be that clots in the venous system of the leg in patients with isolated DVT, are more resistant to embolisation due to different clot structure. Indeed, Undas and colleagues have shown that fibrin clots obtained from PE patients were more permeable, less compact, and lysed more efficiently compared with those with isolated DVT [28]. Furthermore, clot morphology and embolic potential may differ also according to the triggering factor [20], and this might partly explain the reason for a low incidence of asymptomatic PE in our patients with injury or operation. A number of biomarkers have been evaluated in patients with venous thromboembolism. D-dimer levels have been confirmed as the marker of prognosis and extent of embolic disease [29–31]. Elevated levels of troponin were associated with a 5-fold mortality in patients with acute symptomatic PE [11]. Interestingly, it was shown that hsCRP levels are higher in patients with idiopathic compared to secondary venous thromboembolism [32]. However, association between CRP levels and future venous thromboembolic events failed to be demonstrated [33,34]. Increased levels of D-dimer, hsCRP and t-PA were found in our patients with asymptomatic PE, indicating activation of coagulation and inflammatory response, and perhaps also endothelial dysfunction. However, in multivariate logistic regression analysis, the frequency of silent PE was independently associated with the presence of unprovoked DVT only, emphasising again the importance of the DVT etiology. The frequency of asymptomatic PE in patients with thrombophilia has not been assessed yet. In our study, 68.7 % of patients with newly diagnosed antiphospholipid syndrome had asymptomatic PE. Their risk of asymptomatic PE was 4.5-times higher than in other patients. In the cohort of patients with antiphospholipid syndrome, Carvera and colleagues found DVT as the most common presenting clinical manifestation, whereas symptomatic PE was less frequent [35]. However, our results showed that the majority of the patients with antiphospholipid syndrome, clinically presenting with DVT, also have PE. This suggests that PE, especially silent, may be an important clinical manifestation of antiphospholipid syndrome reflecting the systemic prothrombotic state in those patients. Screening for the presence of antiphospholipid syndrome may be useful. Factor V Leiden and prothrombin G20210A mutations did not affect the frequency of asymptomatic PE significantly. Furthermore, Factor V Leiden seemed to be less frequent in patients with asymptomatic PE. This is in concordance with the results of Rossi
and colleagues who studied the influence of thrombophilia on symptomatic PE [36]. The importance of natural coagulation inhibitor deficiency could not be assessed in our study due to the small number of such patients. Asymptomatic PE should not be underestimated. The risk of recurrent PE is increased in patients with silent PE [4–7], and Tzoran and colleagues have shown that 40 % of recurrent PE events were fatal [8]. Therefore, the balance of benefit and risk of long-term anticoagulation might be different in patients with PE or DVT and there has been some debate over the benefit of routine screening for silent PE in patients with DVT. Our study showed the highest incidence of silent PE in patients with unprovoked proximal DVT. Whether screening for PE in this subgroup of patients might be cost-beneficial should be established in prospective studies. Our study has some potential limitations. The patients received at least one therapeutic dosage of low molecular weight heparin before the blood was collected for determination of the biomarkers and this might have some influence on the results. The number of patients included was relatively low for assessing the association between the biomarkers and silent PE in the multivariate statistical models. Furthermore, the small number of the patients with confirmed thrombophilia also limits a firm conclusion on association between thrombophilia and silent PE. A larger number of patients might influence the results of multivariate analysis, and potentially elucidate the importance of thrombophilia, including antiphospholipid syndrome in silent PE. Ventilation-perfusion scintigraphy was used for PE diagnosis in this study, and therefore some patients may have had a misclassification of PE diagnosis due to limitations in the diagnostic accuracy of this method. However, a similar rate of silent PE was observed in a study by Li and colleagues, where CT angiography was used to confirm PE [10]. In conclusion, we demonstrate that patients with unprovoked proximal DVT have a high risk of asymptomatic PE. Whether aggressive and prolonged anticoagulant therapy has a more favourable outcome in patients with detected silent PE should be assessed in prospective clinical trials. Acknowledgments This study was supported by the Slovenian Ministry of Science and Technology, grant No. P3-0308. References [1] White RH. The epidemiology of venous thromboembolism. Circulation 2003;107: I4–8. [2] Kearon C. Natural history of venous thromboembolism. Circulation 2003;107:22–30. [3] Stein PD, Matta F, Musani MH, Diaczok B. Silent pulmonary embolism in patients with deep venous thrombosis: a systematic review. Am J Med 2010;123:426–31. [4] Cuppini S, Cattelan AM, Casara D, Prandoni P. Occult pulmonary embolism in patients with proximal deep venous thrombosis. Ann Ital Med Int 1991;6:1–5. [5] Meignan M, Rosso J, Gauthier H, Brunengo F, Claudel S, Sagnard L. d'Azemar P, Simonneau G, Charbonnier B. Systematic lung scans reveal a high frequency of silent pulmonary embolism in patients with proximal deep venous thrombosis. Arch Intern Med 2000;160:159–64. [6] Monreal M, Büller H, Lensing AW, Bonet M, Roncales J, Muchart J, et al. Should patients with deep vein thrombosis alone be treated as those with concomitant asymptomatic pulmonary embolism? A prospective study. Thromb Haemost 2002;88:938–42. [7] Jünger M, Diehm C, Störiko H, Hach-Wunderle V, Heidrich H, Karasch T, et al. Mobilization versus immobilization in the treatment of acute proximal deep venous thrombosis: a prospective, randomized, open, multicentre trial. Curr Med Res Opin 2006;22:593–602. [8] Tzoran I, Saharov G, Brenner B, Delsart D, Román P, Visoná A, et al. Monreal M; RIETE Investigators. Silent pulmonary embolism in patients with proximal deep vein thrombosis in the lower limbs. J Thromb Haemost 2012;10:564–71. [9] Douketis JD, Kearon C, Bates S, Duku EK, Ginsberg JS. Risk of fatal pulmonary embolism in patients with treated venous thromboembolism. JAMA 1998;279:458–62. [10] Li F, Wang X, Huang W, Ren W, Cheng J, Zhang M, et al. Risk factors associated with the occurrence of silent pulmonary embolism in patients with deep venous thrombosis of the lower limb. Phlebology 2013 [Epub 2013 May 9]. [11] Becattini C, Vedovati MC, Agnelli G. Prognostic value of troponins in acute pulmonary embolism: a meta-analysis. Circulation 2007;116:427–33. [12] Aujesky D, Roy PM, Guy M, Cornuz J, Sanchez O, Perrier A. Prognostic value of D-dimer in patients with pulmonary embolism. Thromb Haemost 2006;96:478–82.
A. Boc et al. / Thrombosis Research 133 (2014) 1011–1015 [13] Venetz C, Labarère J, Jiménez D, Aujesky D. White blood cell count and mortality in patients with acute pulmonary embolism. Am J Hematol 2013;88:677–81. [14] Miyakis S, Lockshin MD, Atsumi T, Branch DW, Brey RL, Cervera R, et al. International consensus statement on an update of the classification criteria for definite antiphospholipid syndrome (APS). J Thromb Haemost 2006;4:295–306. [15] Bajc M, Neilly JB, Miniati M, Schuemichen C, Meignan M, Jonson B. EANM guidelines for ventilation/perfusion scintigraphy: Part 2. Algorithms and clinical considerations for diagnosis of pulmonary emboli with V/P (SPECT) and MDCT. Eur J Nucl Med Mol Imaging 2009;36:1528–38. [16] R Core Team. R: A language and environment for statistical computing. Vienna: R Foundation for Statistical Computing. http://www.R-project.org; 2012 . [accessed 1 August 2013]. [17] Dorfman GS, Cronan JJ, Tupper TB, Messersmith RN, Denny DF, Lee CH. Occult pulmonary embolism: a common occurrence in deep venous thrombosis. Am J Roentgenol 1987;148:263–6. [18] Huisman MV, Büller HR, ten Cate JW, van Royen EA, Vreeken J, Kersten MJ, et al. Unexpected high prevalence of silent pulmonary embolism in patients with deep venous thrombosis. Chest 1989;95:498–502. [19] Monreal M, Rey-Joly Barroso C, Ruiz Manzano J, Salvador Tarrason R, Lafoz Naval E, Viver Manresa E. Asymptomatic pulmonary embolism in patients with deep vein thrombosis. Is it useful to take a lung scan to rule out this condition? J Cardiovasc Surg 1989;30:104–7. [20] Monreal M, Ruíz J, Olazabal A, Arias A, Roca J. Arias A, Roca J. Deep venous thrombosis and the risk of pulmonary embolism. A systematic study. Chest 1992;102:677–81. [21] Moser KM, Fedullo PF, LitteJohn JK, Crawford R. Frequent asymptomatic pulmonary embolism in patients with deep venous thrombosis. JAMA 1994;271:223–5. [22] Nielsen HK, Husted SE, Krusell LR, Fasting H, Charles P, Hansen HH. Silent pulmonary embolism in patients with deep venous thrombosis. Incidence and fate in a randomized, controlled trial of anticoagulation versus no anticoagulation. J Intern Med 1994;235:457–61. [23] Lusiani L, Visonà A, Bonanome A, Pesavento R, Zanco P. The characteristics of the thrombi of the lower limbs, as detected by ultrasonic scanning, do not predict pulmonary embolism. Chest 1996;110:996–1000. [24] López-Beret P, Pinto JM, Romero A, Orgaz A, Fontcuberta J, Oblas M. Systematic study of occult pulmonary thromboembolism in patients with deep venous thrombosis. J Vasc Surg 2001;33:515–21.
1015
[25] Monreal M, Ruiz J, Fraile M, Bonet M, Davant E, Muchart J, et al. Prospective study on the usefulness of lung scan in patients with deep vein thrombosis of the lower limbs. Thromb Haemost 2001;85:771–4. [26] Jiménez D, Díaz G, Marín E, Vidal R, Sueiro A, Yusen RD. The risk of recurrent venous thromboembolism in patients with unprovoked symptomatic deep vein thrombosis and asymptomatic pulmonary embolism. Thromb Haemost 2006;95:562–6. [27] Moser KM, LeMoine JR. Is embolic risk conditioned by location of deep venous thrombosis? Ann Intern Med 1981;94:439–44. [28] Undas A, Zawilska K, Ciesla-Dul M, Lehmann-Kopydłowska A, Skubiszak A, Ciepłuch K, et al. Altered fibrin clot structure/function in patients with idiopathic venous thromboembolism and in their relatives. Blood 2009;114:4272–8. [29] Jeebun V, Doe SJ, Singh L, Worthy SA, Forrest IA. Are clinical parameters and biomarkers predictive of severity of acute pulmonary emboli on CTPA? QJM 2010;103:91–7. [30] Grau E, Tenías JM, Soto MJ, Gutierrez MR, Lecumberri R, Pérez JL. Tiberio G; RIETE Investigators. D-dimer levels correlate with mortality in patients with acute pulmonary embolism: Findings from the RIETE registry. Crit Care Med 2007; 35:1937–41. [31] Becattini C, Lignani A, Masotti L, Forte MB, Agnelli G. D-dimer for risk stratification in patients with acute pulmonary embolism. J Thromb Thrombolysis 2012; 33:48–57. [32] Luxembourg B, Schmitt J, Humpich M, Glowatzki M, Dressler D, Seifried E, et al. Cardiovascular risk factors in idiopathic compared to risk-associated venous thromboembolism: A focus on fibrinogen, factor VIII, and high-sensitivity C-reactive protein (hs-CRP). Thromb Haemost 2009;102:668–75. [33] Hald EM, Brækkan SK, Mathiesen EB, Njølstad I, Wilsgaard T, Brox J, et al. Highsensitivity C-reactive protein is not a risk factor for venous thromboembolism: the Tromso study. Haematologica 2011;96:1189–94. [34] Fox EA, Kahn SR. The relationship between inflammation and venous thrombosis. A systematic review of clinical studies. Thromb Haemost 2005;94:362–5. [35] Cervera R, Piette JC, Font J, Khamashta MA, Shoenfeld Y, Camps MT, et al. EuroPhospholipid Project Group. Antiphospholipid syndrome: clinical and immunologic manifestations and patterns of disease expression in a cohort of 1,000 patients. Arthritis Rheum 2002;46:1019–27. [36] Rossi E, Za T, Ciminello A, Leone G, De Stefano V. The risk of symptomatic pulmonary embolism due to proximal deep venous thrombosis differs in patients with different types of inherited thrombophilia. Thromb Haemost 2008;99:1030–4.
