We report a case of leaflet escape in an Edwards-TEKNA bileaflet valve, in the mitral position. The examination findings of the explanted valve are compared with a similar case of leaflet escape in an original Edward-Duromedics prosthesis. Based on our findings alone, it is not certain whether the TEKNA valve continues to have a higher risk for fracture.
Of about 20,000 original Edwards – Duromedics bileaflet valves ( Baxter Healthcare Corp , Santa Ana , CA ) ( models 3160 and 9120 ) implanted from 1982 to 1988 , 12 cases of leaflet embolization from the mitral position were reported [ 1 ] . The prosthesis was subsequently withdrawn from the market and reintroduced after modifications as the revised Edwards – Duromedics or TEKNA valve ( models 3160R , 3200 and 9120R , and 9200 ) in June 1990. In recent years , we had to carry out two emergency reoperations due to leaflet embolization of Duromedics prostheses. We report a case of a fractured TEKNA prosthesis as well as the case of a fractured original Duromedics prosthesis.
Case reports
Case 1
A 37 – year-old man was admitted to our Department of Cardiology , with acute onset of severe pulmonary edema and low cardiac output. He had undergone mitral valve replacement with a TEKNA prosthesis ( model 9120R , size 27 mm ) 3 years earlier. Severe mitral regurgitation was diagnosed. The intraoperative findings confirmed an absent posterior mitral valve leaflet. After complete excision of the valve , the left ventricle was palpated in an attempt to retrieve embolized fragments of the leaflet. However , this was unsuccessful. The old valve was replaced by a 29 – mm Medtronic Hall prosthesis ( Medtronic Inc , Minneapolis , MN ) . A search for the remaining embolized leaflet particles was undertaken. By computed tomography , one particle was localized at the aortic bifurcation and another in the left common iliac artery. Both were removed via a transperitoneal approach. They were found to be nearly identical in size , confirming that the leaflet had been broken in half ( Fig 1A , B ) . All parts of the explanted valve were returned to the manufacturing company. The patient made a good recovery.
(Fig 1. Central fracture of a leaflet ( case 1 ) , showing both fragments.)
Case 2
A 64 – year-old man who had undergone mitral valve replacement with a Duromedics prosthesis ( model 9120 , size 31 mm ) in 1986 was admitted to our Department of Cardiology with signs of fulminant pulmonary edema. Cardiac catheterization demonstrated severe mitral regurgitation and a critical stenosis of the right coronary artery. Intraoperatively , the mitral valve prosthesis showed a missing posterior leaflet. The valve was replaced by a 25 – mm Medtronic Hall prosthesis and the patient received a saphenous vein graft to the right coronary artery. Similar to the first case , no periphery pulse deficit gave us an indication to the location of embolization. A foreign body was located in the left common iliac artery by computed tomography and clearly identified in the proximal external iliac artery by B – mode ultrasonic examination. The leaflet was recovered via a retroperitoneal approach. The fracture had occured at one of the pivot balls ( Fig 2 ) . The patient made an uneventful recovery postoperatively. The explanted valve fragments were again sent to the manufacturing firm for inspection.
(Fig 2. Recovered leaflet from second patient ; fracture in the region of the pivoting system.)
Examination of the explanted valve fragments
Case 1
A region of pyrolytic carbon loss , approximately 2 mm in length and 80 ( mu ) m in depth , was observed on the seating lip , which coincided with the location of the central fracture of the leaflet. This pyrolytic carbon loss appeared to be primarily the result of impact as opposed to cavitation or erosion , which are usually related to material quality and flow.
Case 2
Contrary to case 1 , the second case showed a fracture at the leaflet’s pivoting system. Substantial damage in the form of severe pitting on the corresponding pivot slot edge was present.
Comment
Cavitation has been recognized as the reason for the series of valve failures in the Edwards – Duromedics prosthesis. Typical forms of damage attributed to cavitation are pitting and microcracking , which can later give rise to complete fractures , eg , of leaflets. The original Duromedics valve was shown to be more susceptible to cavitation than other comparable bileaflet valves. This was attributed to high velocities on valve closure , especially in surfaces of the inflow side. Cavitation also seems to be promoted by the seating lip , found in Duromedics valves , which halts leaflet movement and can cause high – velocity pressure jets [ 2 and 3 ] . The experience gained from the original Duromedics valve led to several modifications , especially aimed at reducing the valve – closing velocity and the forces of cavitation. The radius of the seating lip was changed , a shock absorber was integrated into the sewing ring , and minor changes of the pivoting joints and the porosity of the material were initiated [ 4 ] . In vitro investigations showed higher cavitation thresholds for the TEKNA valve , comparable with the results for other bileaflet valves [ 5 ] . These findings , however , were not confirmed by all authors. The manufacturing firm reported three cases of leaflet escape for the revised model in 1996 [ 6 ] ; the explanted valves did not show changes typical for cavitation. As a possible reason for valve failure , we must also consider small irregularities in the structure of the material used. Pyrolytic carbon is a very hard but also very brittle substance. Therefore , its surface can be damaged by improper handling , eg , contact with a metallic tool. The currently available clinical data on the TEKNA valve do not show a high risk for fracture. Valve defects , typical for cavitation , have so far not been reported. These early results will have to be confirmed by the clinical course of the coming years.
REFERENCES:
1. Radlick PH. Suspension of marketing Edwards Duromedics bileaflet valve models 3160 and 9120, all sizes. Baxter, Important product information, Santa Ana, CA, 1988:1-2..
2. G. Richard, A. Beavan and P. Strzepa, Cavitation threshold ranking and erosion characteristics of bileaflet heart valve prostheses. J Heart Valve Dis 3 Suppl I (1994), pp. 94-103.
3. G.X. Guo, T.H. Chiang, R.C. Quijano and N.H.C. Hwang, The closing velocity of mechanical heart valve leaflets. Med Eng Phys 16 (1994), pp. 458-464.
4. A. Moritz, W. Klepvtko, S. Rodler et al., Six-year follow-up after heart valve replacement with the Edwards Duromedics bileaflet prosthesis. Eur J Cardiothorac Surg 7 (1993), pp. 84-90.
5. C. Kingsbury, R. Kafesjian, G. Guo et al., Cavitation threshold with respect to dP/dt: evaluation in 29 mm bileaflet, pyrolitic carbon heart valves. Int J Artif Organs 16 (1993), pp. 515-520.
6. Edwards TEKNA bileaflet valve clinical report, June 1996. Baxter Healthcare Corporation, Santa Ana, CA, 1996.. (REcor) Address reprint requests to Dr Hemmer, Sana Herzchirurgische Klinik, Herdweg 2, D-70174 Stuttgart, Germany
Author(s): Hemmer W, Doss M, Hannekum A, Kapfer X.
in: Annals of Thoracic Surgery 2000 Mar;69(3):942-4.
Links:
PMID: 10750795 [PubMed - indexed for MEDLINE]
Reply to Walther and Falk.
August 21, 2003 — dossmedicalWe appreciate the attention that Walther et al. showed in our recent article and would like to point out the following issues in response to their comments. Walther and colleagues suggest that our patient cohort is to small to allow for a meaningful analysis. We disagree! Without employing any statistical tests, any clinician will agree, that there is no clinically relevant difference between a mean gradient of 6.5 mmHg and one of 7.4 mmHg, or an ejection fraction of 64.6% and one of 66.6%. Also, a posterior wall thickness of 1.32 cm or one of 1.26 cm has no clinical relevance for the patient. Medically speaking, we would expect to need at least 15–20% differences between the groups to get a clinical relevance. The differences between the groups however, are in the region of 1–2%. Based on the relatively small sample size, after consulting our statisticians and the statisticians from Edwards Life Sciences, who have helped us design and evaluate our study, we can expect to pick up differences of 20% under this constellation.
With regard to intraoperatively measured annular diameters, we found no relevant differences between the groups (21.6 versus 22.4 mm). The subsequently implanted valve sizes are shown in Table 2 of our article [1]
Walther et al. imply that our article lacks information on indexed differences of left ventricular (LV) mass. We would like to call to their attention that we have separately reported all measurements that run into the equation of determining LV-mass index in Tables 1 and 3 of our article for both groups [1].We feel that the reader can thus better understand that there were no significant differences in the single components, rather than just seeing that the indexed end product did not differ.
Walther et al. go on to cite their article in which they report a significantly enhanced LV-mass regression after stentless aortic valve replacement [2]. They compared two different types of stentless valves with one stented valve, in groups that were of different sizes (n=49+n=57 versus n=74). Looking at the gradients that they report it appears that a reduction in transvalvular gradient by 1.4 mmHg (from postop 18.1 mmHg to 16.7 mmHg for stentless valves, with the stented valve gradients remaining at 20.1 mmHg) is statistically significant.
Furthermore, it appears that a pressure difference of 3.4 mmHg is enough to cause a significant reduction in LV-mass hypertrophy.
In our experience these minimal changes in pressure gradients cannot be held accountable for a more complete or faster regression of LV-mass. Our findings were similar to the findings of Cohen et al. from the Cleveland Clinic, who also conducted a prospectively randomized trial comparing stentless with stented aortic valves, and who also found no significant differences in hemodynamic performance or reduction in LV-mass between the groups [3].
In the current environment of science it is, however, not uncommon that different groups, investigating similar projects, come to different conclusions.
References
1. Doss M., Martens S., Wood P., Aybek T., Kleine P., Wimmer-Greinecker G., Moritz A. Performance of stentless versus stented aortic valve bioprostheses in the elderly patient: a prospective randomized trial. Eur J Cardiothorac Surg 2003;23:299-304.
2. Walther T., Falk V., Langebartels G., Krüger M., Bernhardt U., Diegeler A., Gummert J., Autschbach R., Mohr F.W. Prospectively randomized evaluation of stentless versus conventional biological aortic valves: impact on early regression of left ventricular hypertrophy. Circulation 1999;100(19 Suppl):II6-II10
3. Cohen G., Christakis G.T., Joyner C.D., Morgan C.D., Tamariz M., Hanayama N., Mallidi H., Szalai J.P., Katic M., Rao V., Fremes S.E., Goldman B.S. Are stentless valves hemodynamically superior to stented valves? A prospective randomized trial. Ann Thorac Surg 2002;73:767-778.
Author(s): Doss M.
in: Eur J Cardiothorac Surg. 2003 Aug;24(2):335-6.