|Year : 2018 | Volume
| Issue : 2 | Page : 109-115
Clinical profile of patients with prosthetic valve thrombosis treated with fibrinolysis
Jaywant M Nawale, Ajay S Chaurasia, Digvijay Deeliprao Nalawade, Nitinkumar Abdagire
Department of Cardiology, TNMC and BYL Nair Ch. Hospital, Mumbai, Maharashtra, India
|Date of Web Publication||10-Sep-2018|
Dr. Digvijay Deeliprao Nalawade
Department of Cardiology, TNMC and BYL Nair Ch. Hospital, 18-ICCU, Ground Floor, OPD Building, Nair Hospital, Mumbai Central, Mumbai - 400 008, Maharashtra
Source of Support: None, Conflict of Interest: None
Objective: Prosthetic valve thrombosis (PVT) is a potentially life-threatening complication. Fibrinolytic therapy is the first line of treatment in the developing countries mainly due to financial constraints and limited surgical availability. We performed a retrospective, single-center study to evaluate the clinical profile of patients with PVT treated with fibrinolysis. Methods: Data of 21 consecutive patients admitted with PVT and treated with fibrinolysis during 1-year study period (January 2017–January 2018) were retrospectively analyzed in terms of clinical characteristics and outcomes. Results: Out of 21 patients with 21 PVT episodes, there were 13 females (62%) and 8 males (38%) with a mean age of 35.2 ± 12.2 years (age range: 18–57 years). Presenting complaint in all patients was dyspnea with majority (90.5%) presenting in the New York Heart Association (NYHA) Class III/IV. Subtherapeutic international normalized ratio (<2.5) was found in 19 (90.5%) patients. Nineteen (90.5%) PVT episodes occurred in mitral position alone and two (9.5%) in both mitral and aortic. All patients were thrombolyzed with streptokinase with complete clinical success in 16 (76.2%), partial clinical success in 2 (9.5%), and inhospital mortality in 3 (14.3%) patients. Complications occurred in three (14.3%) patients, including one (4.8%) major (embolic stroke) and two (9.5%) minor (per vaginal bleed). Conclusions: Left-sided PVT is more common in mitral position than aortic with subtherapeutic anticoagulation being the major risk factor. Fibrinolysis can be considered as an effective and acceptable alternative to surgery, even for patients presenting with NYHA Class III/IV, especially in developing countries or in centers with limited resources.
Keywords: Fibrinolysis, prosthetic valve thrombosis, streptokinase
|How to cite this article:|
Nawale JM, Chaurasia AS, Nalawade DD, Abdagire N. Clinical profile of patients with prosthetic valve thrombosis treated with fibrinolysis. J Pract Cardiovasc Sci 2018;4:109-15
|How to cite this URL:|
Nawale JM, Chaurasia AS, Nalawade DD, Abdagire N. Clinical profile of patients with prosthetic valve thrombosis treated with fibrinolysis. J Pract Cardiovasc Sci [serial online] 2018 [cited 2019 Dec 15];4:109-15. Available from: http://www.j-pcs.org/text.asp?2018/4/2/109/240960
| Introduction|| |
Thrombosis of prosthetic heart valves is a serious complication with high morbidity and mortality and reported incidence of 0.5%–8% in left-sided mechanical prostheses and up to 20% in tricuspid prostheses. Although rare in the developed world, it occurs frequently in developing countries. In a retrospective study from India, left-sided prosthetic valve thrombosis (PVT) occurred in 6.1% of patients within 6 months of valve replacement.
The treatment options for PVT include surgery, fibrinolysis, and anticoagulation. Due to the lack of larger prospective studies, the choice of treatment for PVT remains controversial. Surgery has been traditionally considered as the first choice for PVT. In our institution and in many other developing countries, thrombolysis is still the treatment of choice because of its ease of availability and the limitations with surgery (long-waiting period and higher cost).
We conducted a retrospective study of consecutive patients with PVT treated with fibrinolysis in a teaching institute in India to study the clinical characteristics and outcomes.
| Methods|| |
This was a single-center, retrospective study with data collected from hospital records of all consecutive patients admitted to our institute during a period of 1 year (January 2017- January 2018). Fibrinolysis was used as a first-line treatment. Clinical profile, investigation records, and outcomes were reviewed. PVT was diagnosed based on clinical presentation and diagnostic procedures, including transthoracic echocardiography (TTE) and fluoroscopy. PVT was considered obstructive if TTE showed a high-pressure gradient across prosthetic valve [Figure 1]a along with reduced or fixed leaflet mobility on fluoroscopy [Figure 2]a and[Figure 2]b. Patients with infective endocarditis and nonobstructive thrombosis or patient-prosthesis mismatch (characterized by high gradients on TTE but normal leaflet motion on fluoroscopy) were excluded from the study. All patients received streptokinase 0.25 mU over 30 min, followed by a 0.1 mU/h infusion over 24 h. Fluoroscopy and TTE were repeated after completing fibrinolysis, and patients with no or incomplete response were referred to surgery.
|Figure 1: Pre- and post-fibrinolysis continuous wave Doppler echocardiographic views in a patient with prosthetic valve thrombosis in mitral position. (a) Prefibrinolysis continuous wave Doppler across mitral prosthesis with peak gradient of 33 mmHg and mean gradient of 25 mmHg suggestive of prosthetic valve thrombosis. (b) Postfibrinolysis continuous wave Doppler across mitral prosthesis with peak gradient of 4 mmHg and mean gradient of 2 mmHg suggestive of normal valve restoration.|
Click here to view
|Figure 2: Pre- and post-fibrinolysis cine fluoroscopy views in a patient with prosthetic valve thrombosis in mitral position, during different phases of cardiac cycle. (a and b) Prefibrinolysis right anterior oblique caudal views showing bileaflet tilting disc valve in mitral position with one leaflet fixed (indicated by arrow) in closed position while the other leaflet in closed position during systole (a) and open position during diastole (b) suggestive of prosthetic valve thrombosis. (c and d) Postfibrinolysis right anterior oblique caudal views showing bileaflet tilting disc valve in mitral position with both leaflets moving to closed position during systole (c) and open position during diastole (d) suggestive of normal valve restoration.|
Click here to view
Postfibrinolysis valve function was considered to be normally restored if the records showed 50% reduction in TTE transvalvular gradients as compared to baseline [Figure 1]b and normal leaflet motion on fluoroscopy [Figure 2]c and[Figure 2]d. “Complete clinical success” of fibrinolytic therapy was considered if there was restoration of normal valve function along with clinical improvement in the absence of death, need for surgery, or any major complication including stroke or major bleeding (intracranial bleed or bleeding requiring transfusion or surgical treatment) during the hospital stay. “Partial clinical success” was defined if restoration of normal valve function was associated with any major complication or incomplete restoration of valve function was achieved (characterized by restricted movement of valve leaflets on fluoroscopy despite normalized transvalvular gradients) which required surgical referral. Death or absence of hemodynamic response (after 24 h of fibrinolysis) was considered as “failure” of fibrinolysis, while mere normalization of transvalvular gradients irrespective of leaflet mobility was considered as “hemodynamic success.”
Statistical analysis was done using SPSS Version 16.0 software (SPSS Inc., Chicago), and data were presented as mean, standard deviation, and percentages.
| Results|| |
Clinical characteristics [Table 1]
During the 1-year study period, 21 patients diagnosed with thrombotic obstructive PVT received fibrinolytic therapy with streptokinase as first-line treatment. Of these patients, 13 were female (62%) and 8 were male (38%) with a mean age of 35.2 ± 12.2 years (age range: 18–57 years) [Table 1]. The mean time between valve replacement and PVT was 21 months (range: 1 month–8 years) [Table 1]. Presenting complaint in all patients was dyspnea, wherein 11 patients (52.4%) were in the New York Heart Association (NYHA) Class IV, 8 (38.1%) in NYHA Class III, and 2 (9.5%) in NYHA Class II [Table 1]. All patients were on warfarin therapy except two who were on acenocoumarol. Subtherapeutic international normalized ratio (INR < 2.5) was found in 19 (90.5%) patients [Table 1] of which two had history of treatment discontinuation, while others had low INR despite therapy, mainly due to irregular INR monitoring.
Of the total 21 PVT episodes, 19 (90.5%) occurred in mitral position alone and 2 (9.5%) in both mitral and aortic positions [Table 1]. Considering the valve types, almost all were bileaflet tilting disc type (ATS-15, St Jude-4, Carbomedics-4) except one single-tilting disc type (TTK CHITRA) [Table 1].
Outcomes of fibrinolysis [Table 2]
The mean gradient across the thrombosed mitral valve ranged from 16 to 43 mmHg with mean of 26.7 ± 7.7 mmHg, while that across thrombosed aortic valve ranged from 25 to 44 mmHg with mean of 34.5 ± 13.4 mmHg [Table 2]. Postfibrinolysis, the mean gradients across mitral valve ranged from 2 to 11 mmHg with mean of 8.2 ± 3.5 mmHg, while that across thrombosed aortic valve ranged from 11 to 16 mmHg with mean of 13.5 ± 3.5 mmHg [Table 2]. Hemodynamic success was achieved in 18 (85.6%) patients [Table 2] using streptokinase with mean dose of 2.65 mU for mean duration of 24 h. Complete clinical success was achieved in 16 (76.2%) patients, of which 2 (100%) were in NYHA Class II and 14 (73.7%) were in NYHA Class III/IV [Table 2]. Three (14.3%) patients died [Table 2] during fibrinolysis due to refractory heart failure and cardiogenic shock. Two patients had restricted leaflet mobility on fluoroscopy despite normalization of transvalvular gradients of which one patient also developed embolic stroke during the hospital stay a day after fibrinolysis. Per vaginal bleeding was found in two patients.
| Discussion|| |
Obstruction of prosthetic heart valves can be caused due to thrombosis, pannus formation, and patient-prosthesis mismatch or rarely due to vegetation. Of these, PVT occurs more frequently in left-sided valves, mitral more than aortic and mechanical more than bioprosthetic type. Risk factors for PVT are inadequate anticoagulation, atrial fibrillation, left ventricular dysfunction, and hypercoagulable states such as pregnancy, early postoperative period, older mechanical prosthesis designs such as ball and cage type, and thrombosis secondary to pannus or vegetation. Presenting features of PVT are shorter duration of symptoms which may include dyspnea developing over weeks to days, insidious onset of fatigue, systemic embolism or sometimes acute pulmonary edema, and even death.
Clinical characteristics of prosthetic valve thrombosis patients
In this study, total 21 patients diagnosed with PVT were treated with fibrinolysis of which majority 13 (62%) were female as compared to 8 (38%) males. Dyspnea was presenting complaint in all patients with majority (90.5%) presenting in NYHA Class III/IV. The proportion is slightly higher than other studies. In institutes where patients are screened regularly by fluoroscopy, PVT can be diagnosed in early stage in NYHA Class I/II (despite normal transvalvular gradients on TTE), while in institutes where fluoroscopy is done, only when patients are symptomatic and/or TTE shows increased transvalvular gradients, majority like here will be diagnosed in Class III/IV. Lack of regular patient follow-up can also lead to delayed presentation in NYHA Class III/IV. Subtherapeutic anticoagulation was the major (90.5%) risk factor associated with PVT which was mainly due to irregular follow-up and poor treatment compliance [Table 1]. Similar observations were also found in studies conducted in developing countries such as Cuba and Nepal where the rates of inadequate anticoagulation were high (75% and 69.6%, respectively)., Other studies from India [Table 3] show variable rates of inadequate anticoagulation ranging from 41.5% to 90.3%. Mitral valve PVT is 2–3 times more frequent than thrombosis in aortic position, which was also found in prior Indian studies [Table 3]. In our study as well, 90.5% of PVT episodes occurred in mitral position alone while 9.5% occurred in both mitral as well as aortic positions [Table 1].
|Table 3: Indian data of patients with prosthetic valve thrombosis treated with fibrinolysis (studies conducted during 1990-2017)|
Click here to view
Diagnosis of prosthetic valve thrombosis
Diagnosis of PVT requires clinical suspicion along with diagnostic investigations such as TTE and transesophageal echocardiography (TEE) showing elevated transvalvular gradients along with reduced or fixed leaflet mobility on echocardiography and/or fluoroscopy. TEE though not used in this study can be additionally helpful to evaluate thrombus burden and differentiate thrombus from pannus, wherein thrombus usually shows homogeneous, mobile, or fixed echo densities located at the valve occluder and/or valve struts, while pannus shows fixed, bright echodense structures, sometimes containing focal calcific deposits, present primarily along the valve ring with extension into the valve orifice. In our study, all patients were diagnosed as PVT based on clinical, TTE, and fluoroscopy findings except one patient with single-tilting disc type (TTK CHITRA) mitral prosthesis in whom clinical presentation and TTE alone were used to diagnose PVT and who also had a rapid downhill course with fatal outcome despite early initiation of fibrinolysis.
Treatment of prosthetic valve thrombosis
Due to lack of strong evidence, the treatment of choice in PVT remains controversial and the guidelines given by various organizations differ. The European Society of Cardiology valve guidelines recommend surgery as a Class I treatment for obstructive valve thrombosis in critically ill patients without serious comorbidity. Thrombolysis should be considered when surgery is not available or patients are unlikely to survive surgery. The Society of Heart Valve Disease recommends thrombolysis as the first choice in all cases of PVT unless it is contraindicated. The American College of Chest Physicians recommends thrombolysis as the first choice for PVT with thrombus burden <0.8 sq.cm. irrespective of NYHA class. Recently, the updated American College of Cardiology/American Heart Association guidelines recommend either low-dose, slow-infusion fibrinolysis or surgery, dependent on multiple factors such as NYHA class, thrombus burden, contraindications to fibrinolysis, surgical risk, and recurrence.
Fibrinolysis and its outcomes
Advantages of fibrinolysis include its simplicity, easy availability, and comparative low cost. In a review by Lim and Lloyd, which involves eight large studies (enrolling ≥50 patients) determining outcome of fibrinolysis in PVT, the success rate with fibrinolytic therapy ranged from 58% to 90% while the 30-day mortality rate ranged from 0.8% to 16.7%. Furthermore, the rate of complications was variable of which thromboembolism ranged from 1.7 to 17.6% while major bleeding ranged from 1.7% to 12%. In the previous Indian studies of patients with PVT treated with fibrinolysis [Table 3], the success rate with fibrinolysis varied from 59% to 100% with overall rate of around 78.1%, while the mortality rate ranged from 1.9% to 9.5% with overall rate of around 5.4%. The rate of major complications (including thromboembolism, major bleed, and anaphylaxis) ranged from 8.4% to 18.7% while minor complications (minor bleed, fever, etc.) ranged from 2.3% to 25%.
In our study, out of 21 patients with 21 PVT episodes, complete clinical success was achieved in 16 (76.2%), partial clinical success in 2 (9.5%), and hemodynamic success in 18 (85.6%) with inhospital mortality of 3 (14.3%) patients [Table 2]. All three patients who died presented with NYHA Class IV. As compared to other studies, the relatively high mortality rate could be due to majority (90.5%) of patients presenting in NYHA Class III/IV and small sample size. Studies by Vasan et al. and Rajasekhar et al. [Table 3] also had smaller sample size, but the patients presenting in NYHA Class III/IV in these studies were 75% (12 of 16) and 61.5% (8 of 13), respectively. Complications occurred in three (14.3%) patients including one (4.8%) major (embolic stroke) and two (9.5%) minor (per vaginal bleed) [Table 2] which were comparable with previous studies [Table 3]. Both patients with partial clinical success showed improvement in functional class after fibrinolysis. Of these two patients, one was referred to redo-surgery and the other due to associated embolic stroke was managed conservatively. The patient with embolic stroke also had associated atrial fibrillation and received appropriate neurological treatment. Patients with per vaginal bleeding were treated conservatively.
Considering the Indian data, majority of Indian studies [Table 3] (except studies by Karthikeyan et al. and Balasundaram et al.) have concluded fibrinolysis as effective alternative to surgery for treating PVT which was also found in our study. Newer fibrinolytic agent, tenecteplase, was also equally efficacious for treating PVT according to the studies by Sharma et al. and Kumar et al. while slow-infusion tenecteplase (0.5 mg/kg intravenously over 24 h) was more effective than streptokinase in treating mitral mechanical PVT according to Kathirvel et al. However, a randomized controlled study conducted by Karthikeyan et al. compared an accelerated infusion with the conventional infusion of streptokinase in 120 patients with a first episode of left-sided PVT and concluded that fibrinolysis was less efficacious (efficacy rate 59%) in treating left-sided PVT and may not be a reasonable alternative to surgery, especially in NYHA Class III/IV patients. To the contrary, in our study, although the data were observational and retrospective, complete clinical success of 73.7% [Table 2] was achieved in patients presenting with NYHA Class III/IV. In addition, Balasundaram et al. in a study comparing the outcome of fibrinolysis in patients with recurrent PVT with those treated for a first episode (51 patients with 75 PVT episodes) and Kumar et al. in a subset of patients (16 patients of total 95) with recurrent PVT found that fibrinolysis was not as efficacious for a recurrent episode of PVT as it was for the first episode. However, in our study, there was no patient with recurrent PVT.
As compared to fibrinolysis, surgery is associated with high mortality rate ranging from 5% to 36% with functional class as an important predictor of the surgical risk. Immediate surgical nonavailability and financial constraints are other limitations of surgery, especially in developing countries. Surgery can be advantageous, especially in patients with recurrent thrombosis, undetermined etiology of obstruction, fibrinolysis failure, or contraindications to fibrinolysis such as large left atrial thrombus >5 mm, ischemic stroke in first 4–6 weeks, and history of intracranial bleed., In a systematic review and meta-analysis, Karthikeyan et al. compared urgent surgery with fibrinolysis for treating left-sided PVT by including seven studies with 690 episodes of PVT, of which 446 were treated with surgery and 244 with fibrinolysis. They concluded that urgent surgery was not superior to fibrinolysis at restoring valve function but substantially reduced the occurrence of thromboembolic events, major bleeding, and recurrent PVT, and in experienced centers, urgent surgery should probably be preferred over fibrinolysis for treating left-sided PVT. Thus, further larger prospective studies comparing the two modalities might be required to decide the optimum choice of treatment while treating PVT.
This was a single-center, retrospective study which included relatively small number of patients. Long-term follow-up outcomes and extended courses of fibrinolysis beyond 24 h were not considered. Recent studies have shown that the use of slow- and low-dose fibrinolysis along with TEE can further increase the efficacy of fibrinolysis,, which was not considered in this study.
| Conclusions|| |
Left-sided PVT is more common in mitral position than aortic. Subtherapeutic anticoagulation is a major risk factor for developing PVT which can be prevented by regular follow-up, prothrombin time-INR monitoring, and ensuring proper treatment compliance. In patients with left-sided PVT, fibrinolysis unless contraindicated can be considered as an effective and acceptable alternative to surgery even in patients with NYHA Class III/IV, especially in developing countries or in centers with limited resources, where multiple factors such as surgical availability, financial cost, and operative mortality are to be weighed while deciding the treatment.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Özkan M, Gündüz S, Biteker M, Astarcioglu MA, Çevik C, Kaynak E, et al
. Comparison of different TEE-guided thrombolytic regimens for prosthetic valve thrombosis: The TROIA trial. JACC Cardiovasc Imaging 2013;6:206-16.
Karthikeyan G, Senguttuvan NB, Joseph J, Devasenapathy N, Bahl VK, Airan B, et al
. Urgent surgery compared with fibrinolytic therapy for the treatment of left-sided prosthetic heart valve thrombosis: A systematic review and meta-analysis of observational studies. Eur Heart J 2013;34:1557-66.
Talwar S, Kapoor CK, Velayoudam D, Kumar AS. Anticoagulation protocol and early prosthetic valve thrombosis. Indian Heart J 2004;56:225-8.
Cáceres-Lóriga FM, Morais H. Thrombotic obstruction in left-side prosthetic valves: Role of thrombolytic therapy. Indian Heart J 2015;67 Suppl 2:S10-2.
Bonou M, Lampropoulos K, Barbetseas J. Prosthetic heart valve obstruction: Thrombolysis or surgical treatment? Eur Heart J Acute Cardiovasc Care 2012;1:122-7.
Balasundaram RP, Karthikeyan G, Kothari SS, Talwar KK, Venugopal P. Fibrinolytic treatment for recurrent left sided prosthetic valve thrombosis. Heart 2005;91:821-2.
Cáceres-Lóriga FM, Pérez-López H, Morlans-Hernández K, Facundo-Sánchez H, Santos-Gracia J, Valiente-Mustelier J, et al
. Thrombolysis as first choice therapy in prosthetic heart valve thrombosis. A study of 68 patients. J Thromb Thrombolysis 2006;21:185-90.
Hirachan A, Roka M, Prajapati D, Adhikari CM, Bishal KC, Khadka MB, et al
. Prosthetic valve thrombosis in a tertiary cardiac centre. Nepal Heart J 2017;14:9-11.
Vasan RS, Kaul U, Sanghvi S, Kamlakar T, Negi PC, Shrivastava S, et al
. Thrombolytic therapy for prosthetic valve thrombosis: A study based on serial Doppler echocardiographic evaluation. Am Heart J 1992;123:1575-80.
Reddy NK, Padmanabhan TN, Singh S, Kumar DN, Raju PR, Satyanarayana PV, et al
. Thrombolysis in left-sided prosthetic valve occlusion: Immediate and follow-up results. Ann Thorac Surg 1994;58:462-70.
Rajasekhar D, Balakrishnan KG, Venkitachalam CG, Tharakan JA, Titus T, Pillai VR, et al
. Thrombolytic therapy for prosthetic cardiac valve thrombosis. Indian Heart J 1994;46:101-5.
Agrawal D, Dubey S, Saket B, Bhargava M, Mehta N, Lohchab SS, et al
. Thrombolytic therapy for prosthetic valve thrombosis in third world countries. Indian Heart J 1997;49:383-6.
Gupta D, Kothari SS, Bahl VK, Goswami KC, Talwar KK, Manchanda SC, et al
. Thrombolytic therapy for prosthetic valve thrombosis: Short- and long-term results. Am Heart J 2000;140:906-16.
Karthikeyan G, Math RS, Mathew N, Shankar B, Kalaivani M, Singh S, et al
. Accelerated infusion of streptokinase for the treatment of left-sided prosthetic valve thrombosis: A randomized controlled trial. Circulation 2009;120:1108-14.
Sharma V, Singh R, Mishra R, Arora AP, Gupta LC, Yadava OP, et al
. Use of tenecteplase for left-sided prosthetic valve thrombosis. J Assoc Physicians India 2012;60:55-8.
Kumar BM, Paul GJ, Swaminathan N, Venkatesan S, Hussain GZ, Gnanavelu G, et al
. Assessment of hemodynamic and clinical response in thrombolytic therapy for prosthetic valve thrombosis. Indian Heart J 2017;69:S6.
Kathirvel D, Paul GJ, Palanisamy G, Gnanavelu G, Ravishankar G, Swaminathan N, et al
. Tenecteplase versus streptokinase thrombolytic therapy in patients with mitral prosthetic valve thrombosis. Indian Heart J 2017. [Doi: org/10.1016/j.ihj. 2017.10.014].
Ozkan M, Kaymaz C, Kirma C, Sönmez K, Ozdemir N, Balkanay M, et al
. Intravenous thrombolytic treatment of mechanical prosthetic valve thrombosis: A study using serial transesophageal echocardiography. J Am Coll Cardiol 2000;35:1881-9.
Joint Task Force on the Management of Valvular Heart Disease of the European Society of Cardiology (ESC), European Association for Cardio-Thoracic Surgery (EACTS), Vahanian A, Alfieri O, Andreotti F, Antunes MJ, et al
. Guidelines on the management of valvular heart disease (version 2012). Eur Heart J 2012;33:2451-96.
Lengyel M, Horstkotte D, Völler H, Mistiaen WP, Working Group Infection, Thrombosis, Embolism and Bleeding of the Society for Heart Valve Disease. Recommendations for the management of prosthetic valve thrombosis. J Heart Valve Dis 2005;14:567-75.
Salem DN, O'Gara PT, Madias C, Pauker SG. Valvular and structural heart disease: American College of Chest Physicians evidence-based clinical practice guidelines (8th
edition). Chest 2008;133:593S-629S.
Nishimura RA, Otto CM, Bonow RO, Carabello BA, Erwin JP 3rd
, Fleisher LA, et al
. 2017 AHA/ACC focused update of the 2014 AHA/ACC guideline for the management of patients with valvular heart disease: A Report of the American College of Cardiology/American Heart Association task force on clinical practice guidelines. Circulation 2017;135:e1159-95.
Lim WY, Lloyd G, Bhattacharyya S. Mechanical and surgical bioprosthetic valve thrombosis. Heart 2017;103:1934-41.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3]