Microsoft word - thrombocheck final by sagie.doc

Initial clinical experience with hand – held device (Thrombocheck) for the detection of
bileaflet prosthetic valve malfunction.
Ben Zekry S, Sagie A, MD, Ben-Dor I, MD, Nukrian H, ARDS, Battler A, MD, Shapira Y, MD. Echocardiography unit, Department of Cardiology, Rabin Medical Center, Beilinson Campus, Petah Tiqva, affiliated with Sackler Faculty of Medicine, Tel-Aviv University, Tel Aviv, Israel. Abstract
Introduction
Early recognition of subclinical prosthetic valve (PV) malfunction may allow for early treatment and avoidance of serious complications. Echocardiography cannot be applied on a daily basis. Therefore a hand – held device (Thrombocheck) capable of detecting subtle changes in the acoustic sounds of prosthetic valve has been developed for routine home monitoring of heart valve function. We report our initial clinical experience with this device. Seventy one consecutive patients with one or more bileaflet prosthetic mechanical valves at any position were assessed both by transthoracic echocardiography (TTE), fluoroscopy and by Thrombocheck. These patients attended our clinic for either routine echocardiography (62 patients) and for detection of prosthetic valve malfunction (9 patients). The Thrombocheck was held for one minute in subxhiphoid position perpendicular to the patient, and provided any of the 3 indications: OK – normal function of the PV; Warning – after 3 repeat measurements (1 min each) - abnormal function of the PV; No signal – the not identified The study group had in total 82 bileaflet valves: 47 mitral, 31 aortic and 4 tricuspid. There were 8 (10%) patients with "no signal" indication. Of the remaining 63 patients, there were 10 patients (18.9%) with WARNING alarm in whom 8 had echocardiographic and fluoroscopic evidence of current abnormal leaflet motion, one patient has a recent history of abnormal leaflet motion, and one patient had no evidence of prosthetic valve malfunction. The sensitivity and specificity for detecting abnormal PV malfunction were 90% and 98%, respectively. Conclusion
Thrombocheck had an excellent sensitivity and specificity for the detection of prosthetic valve malfunction in a cohort of patients with bileaflet mechanical prosthetic heart valves. Mechanical prosthetic heart valves carry a yearly complication rate of 2-3%, including malfunction due to thrombosis and tissue ingrowth (1). These complications are associated with high morbidity and mortality rate. Early recognition of subclinical prosthetic valve malfunction may allow early treatment and avoidance of serious complications (hemodynamic instability or thromboembolism). Echocardiography, which is used for routine follow up of patients with prosthetic valves, cannot be applied on a daily basis and requires skilled echocardiographers. Several reports showed typical pattern of closing of the prosthetic valves on real-time sound spectroanalysis at normal condition and with dysfunction of the prosthetic valves (2-5). Fritzsche et al showed that the acoustic sounds of the prosthetic valve remain constant under similar conditions and that patients’ valve signals didn’t change significantly over 6 month period (6). Animal studies showed that frequent control of the heart valve sounds can detect very subtle changes in the integrity of the heart valve sounds before they lead to hemodynamic or thromboembolic consequences, readily detectable by echocardiography (7). Therefore a hand – held device (Thrombocheck) has been developed for routine home monitoring of heart valve function. The individual signature of the heart valve sound is computed from its acoustic information. It detects valve acoustic sounds in the region of 11 kHZ, and is able to detect subtle changes in the acoustic sounds. We report the results of routine use of this new hand – held device in clinical practice. Seventy one consecutive patients with at least one bileaflet prosthetic mechanical valve at any position who attended our clinic for either routine echocardiography or for detection of valve malfunction, were assessed both by echocardiography and by hand–held device (Thrombocheck). Transthoracic echocardiography was done with Sonos 5500 equipment (Philips, Andover, MA, USA) with second harmonic capabilities. All the prosthetic valves were carefully assessed both for hemodynamics and for leaflet motion in multiple views, as previously reported (8). Transesophageal echocardiography (TEE) was done as needed with the same echocardiographic machine and a multiplane 3.7/5 MHz probe. Prosthetic valve malfunction was suspected when echocardiography showed high gradients across the prosthetic valve or increased (>25%) gradient across the prosthetic valve comparing the past examination, or inability to demonstrate full range motion of the two discs. In any case of suspicion of valve dysfunction the patient was referred for fluoroscopy and/or TEE. There were no exclusion criteria for testing the Thrombocheck. In patients with 2 prosthetic valves with an immobilized leaflet in one of them, the analysis of diagnostic accuracy of the device was applied to this very valve only. In the analysis of patients after mitral valve replacement we included only the double disc mitral valve. The Thrombocheck was set for double disc sounds. We used the same Thrombocheck The Thrombocheck was held for one minute in subxhiphoid position perpendicular to the patient, which was lying in a recumbent position. There were 3 optional textual results on the OK – indicated normal function of the prosthetic valve. Warning – after 3 repeat measurements (1 min each) indicated abnormal function of the No signal – the Thrombocheck did not identify valve sounds. Seventy one patients were enrolled (25 male, 46 female), aged 58.5 ± 13.3 years old (range 25 years – 82 years). These patients had in total 82 bileaflet valves: 47 mitral, 31 aortic and 4 The annular diameters of the mitral, aortic and tricuspid valves ranged from 25-31 mm (median 27 mm), 19-25 mm (median 21 mm), and 27-33 mm, respectively. The valve models included were St Jude Medical (St. Paul, Minnesota, USA), CarboMedics (Sulzer Carbomedics, Austin, Tx) and Sorin Bicarbon (Sorin Biomedica, Saluggia, Italy). There were 8 patients with "No signal" indication (one of them was after isolated TVR). Out of the remaining 63 patients, there was a WARNING alarm in 10 (18.9%) with the following clinical and echocardiographic findings: 8 patients had evidence of current abnormal leaflet motion (Table 2); one patient had no evidence of prosthetic valve malfunction, and one patient was two months after thrombolysis therapy due to valve dysfunction and high gradient across the prosthetic valve. This patient was readmitted several months later with obstructive valve thrombosis, and was reoperated. The operative specimen showed combined thrombus and pannus formation. Six patients with abnormal leaflet motion were tested with the device following anti-thrombotic or thrombolytic treatment (Table 2). In 4 patients with successful thrombolysis, the Thrombocheck showed normal PV function ("OK"). In one patient with anti-thrombotic treatment and chronic abnormal leaflet motions there was a “No signal” indication. In one patient with recurrent hospital admissions because of abnormal leaflet motions and known high pressure gradient on the mitral valve due to prosthesis-patient mismatch, the Thrombocheck showed WARNING after treatment, whereas the fluoroscopy showed normal discs movement, and the pressure gradient across the mitral valve returned to baseline. There was one patient after tricuspid valve replacement with chronic abnormal leaflet motion in whom the Thrombocheck indicated OK. Table 3 summarizes the testing results of the Thrombocheck and the impact of valve position on its diagnostic accuracy. There were only one false positive test and one false negative test. After exclusion patients with ”No signal” indication, the Thrombocheck had a sensitivity and specificity of 90% and 98%, respectively, for the detection of prosthetic valve malfunction. The subgroup analysis according to valve position indicates sensitivity of 100% and specificity of 97% for patients with MVR and 100% specificity and 95% sensitivity for patients with AVR. In the current study the hand – held - Thrombocheck had an excellent sensitivity and specificity (90% and 98%) for the detection of prosthetic valve malfunction in a cohort of patients with bileaflets prosthetic heart valves, both for the mitral and the aortic position. A high diagnostic accuracy is an invaluable prerequisite for the daily implementation of the thrombocheck. We should not miss valve malfunction at one hand, and false alarm indications are undesirable as well. The high sensitivity of the thrombocheck provides assurance for both Two of our patients demonstrate the usefulness of the Thrombocheck. In one patient (number 8 in Table 2) the Thrombocheck provided various indications which paralleled with the leaflet motion: it showed WARNING early after thrombolysis, and then OK after 7 months. A few months later it could not be calibrated for this very patient for bileaflet valve, and a repeated obstructive valve thrombosis was diagnosed. The other patient (number 9 in Table 2) with moderately elevated aortic valve gradients (peak 50 mmHg, mean 30 mmHg) in a 19 mm CarboMedics prosthesis, the Thrombocheck indicated WARNING. Fluoroscopy showed only 30 degree limitation of the combined travel angle of the valve. These examples represent the high sensitivity of the Thrombocheck for the detection valve malfunction, even in asymptomatic patients. This early recognition of subclinical prosthetic valve malfunction may allow early treatment before development of serious complications which may lead to hemodynamic instability, thromboembolism and death. It may also detect valve thrombosis at a stage where the thrombus burden will not be associated with high complication rate when Follow up of patients with valve malfunction who were treated (by either antithrombotic therapy or intensified anticoagulation) showed resolution of the WARNING indication in four patients. These resolutions of the WARNING indication further strengthens the reliability of the Thrombocheck. The device may be especially suitable for patients with a prior history of prosthetic valve malfunction. These patients have a higher likelihood of additional episodes, Thrombocheck is planned for individual use and a previous study reported the use of individual devices for patients without reporting any difficulty in detecting valve signal (7). In 10% of our patients the Thrombocheck indicated “No signal”. We do not know if this indication reflects the limitation of the Thrombocheck, valve dysfunction, or (probably) the use of the same hand–held device (same acoustic fingerprint) for all patients. The number of patients in the current study is limited. In addition, we used the same hand – held device for all patients. It is expected that if patients receive a personal device, and this device will initially be checked for appropriate valve function, the accuracy of the device will further improve. Moreover, those patients who will initially show "no signal" will not be offered We report, for the first time, initial experience with the hand – held device - Thrombocheck in the clinical practice. It is a simple and comfortable device for home monitoring with an excellent sensitivity and specificity for the detection of prosthetic valve malfunction. Nevertheless, we believe that it should serve as a screening tool, for assurance at one hand and alert at the other hand, but it cannot replace echocardiography or fluoroscopy in the thorough Vesey JM, Otto CM. Complications of prosthetic heart valves. Curr Cardiol Rep. 2004 Ebner E, Sehner J, Eger H, Ursinus W. Soud spectrographic studies in the diagnosis of dysfunctions of mechanical heart valve prostheses (Bjork-Shiley prosthese). Z Kardiol.1983 Eberhardt AC, Chassaing CE, Ward MA, Lewandowski SJ. Acoustic characterization of mechanical valve condition and loading. J Heart Valve Dis. 1995 Nov;4(6):649-58. Kim SH, Lee HJ, Chang BC. Spectral analysis of heart valve sound for detection of prosthetic valve diseases. Yonsei Med J. 1998 Aug;39(4):302-8. Reynolds KJ, Stephen RO. Acoustic analysis of the closing sounds of implanted prosthetic heart valves. J Acoust Soc Am. 1995 Jul;98(1):69-77. Fritzsche D, Eitz T, Grimmig O, Brensing A, Minami K, Korfer R. Home monitoring of patients after prosthetic valve surgery - experimental background and first clinical attempts. Thorac Cardiovasc Surg. 2004 Aug;52(4):211-7. Eitz T, Fritzsche D, Grimmig O, Frerichs I, Frerichs A, Hellige G, Minami K, Korfer R. Acoustic phenomena and valve dysfunction in cardiac prostheses: data acquisition and collection via the Internet. J Heart Valve Dis. 2003 Jul;12(4):414-9. Shapira Y, Vaturi M, Perlmuter M, Sagie A. Feasibility of 2-dimensional bileaflet valve imaging: a prospective transthoracic echocardiographic study. J Heart Valve Dis 2002; 11: 576- Tong AT, Roudaut R, Ozkan M, Sagie A, Shahid MS, Pontes Junior SC, Carreras F, Girard SE, Arnaout S, Stainback RF, Thadhani R, Zoghbi WA; Prosthetic Valve Thrombolysis- Role of Transesophageal Echocardiography (PRO-TEE) Registry Investigators. Transesophageal echocardiography improves risk assessment of thrombolysis of prosthetic valve thrombosis: results of the international PRO-TEE registry. J Am Coll Cardiol. 2004; 43: 77-84. 10. Gupta D, Kothari SS, Bahl VK, Goswami KC, Talwar KK, Manchanda SC, Venugopal P. Thrombolytic therapy for prosthetic valve thrombosis: short- and long-term results. Am Heart J. AVR – aortic valve replacement, MVR – mitral valve replacement, TVR – tricuspid *Two patients with single disc mitral valves. ** Two patients with biological tricuspid valves.
Table 2: Characteristic, therapy and outcome of patient’s with abnormal leaflet motions. CM – CarboMedics. NA – Not available. SB – Sorin Bicarbon; SJM - St. Jude
Medical;. * This patient had recurrent abnormal leaflet motions
** OK after 7 months, device could not be calibrated for double disc at later follow-
up – with repeated valve obstruction.
Table 3: Effect of valve position on diagnostic accuracy of the Thrombocheck* All patients
Number of patients
No signal indication (%)
Patients excluded from analysis due to
abnormality in an additional valve
Patients analyzed for diagnostic accuracy
Sensitivity (%)
Specificity (%)
AVR – mitral valve replacement; MVR – mitral valve replacement; TVR
– tricuspid valve replacement.

*The statistical analysis was done after exclusion patients with “No signal” indication.
** The associated abnormal valve was aortic in 2 patients and tricuspid in 1 patient.

Source: http://www.m-tech.co.il/photos/uploads/Thrombocheck%20final%20by%20sagie.pdf