Prosthetic replacement of tricuspid valve: bioprosthetic or mechanical

Prosthetic replacement of tricuspid valve: bioprosthetic or mechanical
of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Share
Transcript
  Prosthetic Replacement of the Tricuspid Valve:Biological or Mechanical? Giulio Rizzoli,  MD  , Lorenza De Perini,  MD  , Tomaso Bottio,  MD  ,Giuseppe Minutolo,  MD  , Gaetano Thiene,  MD  , and Dino Casarotto,  MD Divisions of Cardiac Surgery and Cardiovascular Pathology, University of Padova, Padova, Italy Background . Incidence of tricuspid prosthesis replace-ment was 1.9% of all valvular operations performedbetween June 6, 1966 and April 18, 1996. Many seriesreport similar figures, but institutional experience islimited and the consensus on treatment modalities islacking.  Methods . One hundred tricuspid operations were per-formed on 83 patients (46 female). A primary operationwas performed in 64 cases, 13 patients had one previousoperation, 4 patients had two previous operations, and 2patients had three previous operations. Seventeen pa-tients required a tricuspid prosthetic valve rereplace-ment. There were 2 emergent and 17 urgent operations.The New York Heart Association class was IV in 13patients (mean pulmonary artery pressure, 41 mm Hg),III in 66 patients (mean pressure, 38 mm Hg), and II in 21patients. The most frequent operation was simultaneousreplacement of the mitral and tricuspid valve (41 pa-tients). Seventy biological and 30 mechanical prostheseswere used. Total follow-up time was 613 years, mean 7.4years (median 4.2 years), with a maximum of 27.8 years,and was 92% complete. Results . Operative mortality was 24%. Survival was0.54 (0.48 to 0.59, n  39) at 5 years, 0.38 (0.32 to 0.44, n  27) at 10 years, 0.31 (0.25 to 0.36, n  19) at 15 years, 0.29(0.23 to 0.34, n  11) at 20 years, and 0.17 (0.098 to 0.26, n  3) at 25 years. Early mortality was increased from higherNew York Heart Association class (hazard ratio    2.2),congenital disease (hazard ratio  6.9), and valvuloplastyfailure (hazard ratio    4.3). The constant risk phase(4%/patient-year) after 2 years was enhanced by olderoperative age (hazard ratio  1.4). Prosthetic type had noindependent effect. Biological prostheses were at risk for300 years and had a reoperation incidence of 4.7%/ patient-year (14 events); mechanical prosthesis were atrisk for 137 years with a rate of 2.2%/patient-year (3events) (  p  0.21). Three valve thromboses were observedin old-design mechanical prosthesis. Bioprosthetic de-generation showed a steeper rate after 7 years. Conclusions . This study does not show a clear superi-ority of biological versus mechanical prostheses. In thelong run survival with mechanical prostheses could besuperior, given the high rate of bioprosthetic degenera-tion after 7 years.(Ann Thorac Surg 1998;66:S62–7)© 1998 by The Society of Thoracic Surgeons T ricuspid valve replacement is an uncommon opera-tion [1, 2]. Many institutional series involve a smallstudy sample. High perioperative mortality causes areduction in statistical accuracy, and the difficulty indiscriminating mortality from cardiac disease or failureof the tricuspid device make studies of long-term out-come hard to evaluate.The purpose of this study was to evaluate the imme-diate and long-term risk of our series of 100 tricuspidvalve prosthetic replacements between 1966 and 1996and to contribute to the consensus on prosthetic devices. Material and Methods  Patients and Procedures Between June 6, 1966 and April 18, 1996 5,095 valvularoperations were performed at the University of Padova.One hundred were tricuspid valve replacements (1.96%).There were 46 women and 18 men, and the medianoperative age was 45 years (range, 3 to 74 years). Tricus-pid prostheses were inserted in the first open-heartprocedure (64 patients), in the second (13 patients), in thethird (4 patients), and in the fourth (2 patients). Biologicalprosthesis were used in 70 operations and mechanicalprosthesis in 30.The most frequent operation was the concomitantreplacement of the mitral and tricuspid valve in 72patients. The second most frequent procedure was iso-lated replacement of the tricuspid valve in 12 patients.The third was replacement of aortic, mitral, and tricuspidvalves in 9 patients, and the fourth was reoperation onboth mitral and tricuspid prostheses in 7 patients (Table1).In the group of 64 primary operations, 28 patientsunderwent surgery for tricuspid incompetence second-ary to mitral valve disease, 21 for rheumatic disease of thetricuspid valve, 4 for floppy valve disease, 8 for congenitallesions, and 2 for endocarditis. Prosthetic devices of different types were used; they were classified as caged Presented at the VII International Symposium on Cardiac Bioprostheses,Barcelona, Spain, June 13–15, 1997.Address reprint requests to Dr Rizzoli, Cariochirurgia, Via Giustiniani 2,35128 Padova, Italy (e-mail: rzzgli@ux1.unipd.it). © 1998 by The Society of Thoracic Surgeons 0003-4975/98/$19.00Published by Elsevier Science Inc PII S0003-4975(98)01036-4  ball (n    5), caged disc (n    6), tilting disc (n    15),porcine (n  67), pericardial (n  3), and bileaflet (n  4)prostheses (Table 2).The most frequent indication for tricuspid replacement(Table 3) was rheumatic disease (39 patients), followed bybioprosthetic degeneration (13), progression of tricuspidvalve incompetence (10), and congenital disease (8). Theprevalent lesion of the native tricuspid valve was incom-petence (54 patients). Stenosis was present in 2 patientsand stenoinsufficiency in 19. All patients were controlledby telephone interview at a common closing date of June1, 1997. Eight patients, all of whom underwent surgery inthe early years of the study, have not yet been traced. Statistics Proportions are reported with their 70% confidence limits(CI). Continuous variables are summarized as means  1standard deviation. Survival data have been analyzedwith the STATA (STATA Corporation, College Station,TX) and SAS (SAS Institute Inc, Cary, NC) statisticalpackage and displayed by the Kaplan-Meier method. Logrank test has been used to identify differences in survival.Control of confounding and risk factor screening wasdone with the hazard procedure [3] and a stepwisestrategy. Results Overall mortality was 56 patients. The causes of death aredetailed in Table 4; there were 12 unknown late deaths.Operative (30 days) mortality was 24% (16% to 33.6%).Mortality was 26.87% between 1966 and 1980, 18.18%between 1980 and 1990, and 18.8% between 1990 and 1997 (p    NS). Mortality was 25% in 64 patients operated on Table 1. Current Versus Original Operation Current OperationOriginal OperationBiological Prosthesis*** **R *C* *R* *RC *RR C** R** RR* RRC RRR Total**R 8 2 0 2 1 1 0 1 0 1 0 16*CR 0 1 0 0 1 0 0 0 0 0 0 2*RR 24 0 1 4 1 6 0 1 1 0 1 39CRR 1 0 0 0 0 0 0 0 0 0 1 2R*R 0 0 0 0 0 0 0 0 0 0 1 1RCR 1 0 0 0 1 0 0 0 0 0 0 2RRR 6 0 0 0 0 1 1 0 0 0 0 8Total 40 3 1 6 4 8 1 2 1 1 3 70Mechanical Prosthesis*** *CC *RC *RR RCR RRC**R 4 1 1 0 0 0 6*RR 17 0 0 1 0 1 19RCR 0 0 0 1 0 0 1RRR 3 0 0 0 1 0 4Total 24 1 1 2 1 1 30 First symbol position  aorta; second  mitral; third  tricuspid; C  conservative procedure; R  replacement; *  untouched. Table 2. Details of Prostheses Used in This Study Biological Prostheses Mechanical ProsthesesProsthetic Type Freq. Cum. Prosthetic Type Freq. Cum.Hancock formaline 1 1.43 Smeloff 4 13.33Hancock Glutaraldehyde 46 67.14 Pemco 1 16.67Hancock II 7 77.14 Hufnagel 6 36.67Carpentier-Edwards porcine 3 81.43 Lillehei 5 53.33Intact xenotech 1 82.86 Bjork delrin 6 63.33Liotta 6 91.43 Bjork pirolitic 1 76.67St. Jude Medical 1 92.86 Sorin monocast 3 86.67Biocor 2 95.71 St. Jude Medical 4 90.00Ionescu-Shiley 1 97.14Sorin pericarbon 2 100.00Total 70 100.00 30 100.00 S63 Ann Thorac Surg CARDIAC BIOPROSTHESES RIZZOLI ET AL1998;66:S62–7 TRICUSPID VALVE PROSTHESIS  primarily versus 22% in 36 reoperated patients (  p  NS),29% in 62 patients operated for incompetence versus 19%in 21 patient operated for stenosis-stenoinsufficiency(  p    0.37), and 50% in 8 congenital patients versus 50%in 8 patients with valvuloplasty failure (70% CI; 27.4 to72.6). Reoperation mortality was 11.8% (4.0% to 25.6%)(  p    NS). Perioperative deaths were from acute or sub-acute low output syndrome (22 patients), neurologiccomplications (1), and arrythmia (1). One mechanicalassistance and one permanent pacemaker were inserted.Total follow-up time was 613 years, with a mean of 7.4years (median, 4.2 years) and a maximum of 27.8 years.The study was 92% complete. The untraced patients hadcaged ball (7 patients) and tilting disc prostheses (1).Actuarial survival was 0.54 (0.48 to 0.59, n  39) at 5 years,0.38 (0.32 to 0.44, n  27) at 10 years, 0.31 (0.25 to 0.36, n  19) at 15 years, 0.29 (0.23 to 0.34, n  11) at 20 years, and0.17 (0.098 to 0.26, n    3) at 25 years (Fig 1). The hazardfunction shows an early risk phase that merges after 2years with a constant phase of 4.4%/patient-year.Survival was not significantly different (  p    0.12)between mechanical and biological prostheses (Fig 2) andwas not affected by the number of previous operations,number of concurrent valvular procedures, or operativepattern (triple valve replacement, mitroaortic replace-ment, single tricuspid replacement) (Fig 3). It was signif-icantly shortened in congenital patients and patients withvalvuloplasty failure (  p    0.016) (Fig 3D).Multivariable analysis of the risk factors showed thatvalvuloplasty failure (hazard ratio 4.3,  p  0.02), congen-ital disease (hazard ratio 6.9,  p    0.004), and New York Heart Association (NYHA) class (hazard ratio 2.2,  p   0.04) were the only significant factors of the early phaseand older operative age (hazard ratio 1.4,  p    0.03) wasthe only incremental risk factor in the late constantphase.  Fig 1. Tricuspid valve replacement; actuarial versus predicted sur-vival. Actuarial survival  circles, 70% confidence limits  singledash, parametric survival  continuous line. Predicted survival  70% confidence limits  dotted lines. Fig 2. Tricuspid prosthesis. Stratified actuarial survival,  p     2  log rank  p  value. Table 3. Operative Indication Indication Biological Mechanical TotalRheumatic disease 28 11 39Biological prosthesisdegeneration13 0 13Commissurotomyfailure3 2 5Congenital disease 6 2 8Dehiscence 5 2 7Embolism 1 0 1Floppy valve disease 4 1 5Progression of TVincompetence0 10 10Acute endocarditis 2 0 2Unknown 3 0 3Pannus 2 1 3Mechanic structuralfailure0 1 1Thrombosis 3 0 3Total 70 30 100 TV  tricuspid valve. Table 4. Causes of Death Cause Biological Mechanical TotalAcute cardiac failure 2 3 5Congestive heart failure 19 3 22Postoperative hemorrhage 1 0 1Left ventricular outflow tractobstruction1 0 1Acute endocarditis 1 0 1Thrombosis 1 0 1Cerebral embolism 0 2 2Anticoagulant-relatedhemorrhage1 0 1Sepsis 0 1 1 Valvular miocardiopathy 4 0 4Liver failure 2 1 3Cancer 2 0 2Unknown 10 2 12Total 44 12 56 S64  CARDIAC BIOPROSTHESES RIZZOLI ET AL Ann Thorac SurgTRICUSPID VALVE PROSTHESIS 1998;66:S62–7  During follow-up 17 patients required reoperation onthe tricuspid prosthesis. The device was incompetent in 4patients, stenotic in 4, and both stenotic and insufficientin the others. Reoperation concerned the sole tricuspidprosthesis in 3 patients; in all the others it involved themitral valve, the aortic valve, or both. Indications fortricuspid valve replacement were degeneration of bio-prosthetic valve in 8 patients, dehiscence in 1, pannusingrowth in 2, structural failure in 1, valve thrombosis in3, and prophylactic in 2 (Table 5).At 5 years (n    37), 90% (range, 84.5% to 93%) of ourpatients were reoperation-free (Fig 4). The statistics at 10and 15 years were (n  23) 66.5% (range, 57.8% to 73.7%)and (n  5) 52% (range, 41.5% to 61.4%), respectively. Thetotal time at risk was 437.5 years and reoperation inci-dence was 3.9%/patient-year. Biological prostheses wereat risk for 300 years and had a reoperation incidence of 4.7%/patient-year (14 events), mechanical prostheseswere at risk for 137 years, with a rate of 2.2%/patient-year(  p    0.21). Considering the two most representativegroups, it was 4.0%/patient-year over 208 years in por-cine valves and 2.0%/patient-year over 98 years in tiltingdisk valves (not significant). Porcine valve failure steadilyincreased after 7 years. Comment Prosthetic tricuspid valve replacement is often the onlyremaining procedure for patients with endocarditis orcongenital diseases or patients with persistent congestiveheart failure and depressed right ventricular perfor-mance. This explains the low prevalence of prosthetictricuspid valve replacement, which was 1.9% in our seriesof 5,095 operations and 1.8% and 1.7%, respectively, inthe series of Munro and colleagues [2] and McGrath andcoworkers [4]. In the same period we performed 350tricuspid valvuloplasty procedures and only 8 progressedto tricuspid replacement.Selection of high-risk surgical candidates, particularlypatients with congenital disease (Ebstein, common atrio-ventricular orifice, double outlet right ventricle, ventric-ular septal defect) and tricuspid valvuloplasty failure,  Fig 3. Univariate actuarial analysis: (A) Number of concurrent procedures. (B) Most common operative patterns. (C) Open heart procedures.(D) Tricuspid valve operative indication. Stratified actuarial survival,  p     2  log rank  p  value. ARD  acute rheumatic disease; f.   fail-ure; proc.   procedure; RRR  triple valve replacement; *RR  mitrotricuspid replacement; **R  tricuspid replacement; v. deg.  valve de- generation. S65 Ann Thorac Surg CARDIAC BIOPROSTHESES RIZZOLI ET AL1998;66:S62–7 TRICUSPID VALVE PROSTHESIS  explains our 18.8% mortality (since 1980), similar to themortality reported by other authors [1, 5].Duran [6] has pointed out that incompetence of tricus-pid valvuloplasty is related to failure of the left heartcomponents. Holper and associates [7] reported that 6 of 8 patients had concomitant mitral valve or mitral pros-thesis dysfunction. Similarly, 4 of our patients had mitralprosthesis dehiscence. In the Mayo clinic series [8],patients with mitral and aortic prostheses who developedtricuspid incompetence also had a poor prognosis withhigh mortality rates.Conversely, patients with rheumatic tricuspid diseaseand stenosis of the valve had, in our series, a loweroperative mortality (  p  NS). These patients were expe-riencing their first operation and were possibly relievedby an effective valvular function with a similar or inferiorgradient.These patients are further exposed to a constant deathrisk of 4.4% of patients per year, resulting in a long-termsurvival of 31% at 15 years and 29% at 20 years, resultsthat are similar to most large series [1, 9, 10]. The onlysignificant factor related to mortality was older age,which is an obvious confounder of all survival studies.No other variable was significant, probably because of the low power of the study. Biological prostheses (39patients) did not show superior results to mechanicalprostheses (22 patients).Supporters of biological prostheses [11, 12] contendthat these valves do not require anticoagulation and havea slower degenerative evolution than in mitral or aorticposition, because of the low pressure stress in the rightchambers. Massive thrombosis of the biological prosthe-ses are rare and related to blood stasis [13]. This compli-cation was not observed in the large series of Glower andcoworkers [10], the series of Munro and associates [2], orthis series. Guerra and colleagues [12] described pannusformation on the ventricular side of the mural cusp, apossible consequence of stasis and low-grade thromboticapposition.Opponents do not deny the widely recognized prob-lem of prosthetic valve thrombosis with biological pros-theses, but they relate this complication to old prostheticdesign [1]. Survival between prostheses was comparablewith recent devices [14] or, in the long term, was superiorin the mechanical prosthesis [9] because of degradationof bioprosthetic valves. In this series we observed threevalve thromboses affecting exclusively old-design me-chanical valves. We report two instances of sudden deathin patients with Hancock tricuspid valves, but both thesepatients also had tilting disc mitral prostheses.New bileaflet valves are credited with improved hemo-dynamics, low gradients, reduced turbulence, and opti-mal durability, but recently Aoyagi and colleagues, re-ported three patients with thrombosis [15] of the St. JudeMedical valve in the tricuspid position. Conversely, bio-prosthetic degeneration in the tricuspid position mayseverely impair right heart hemodynamics. Degenerationand leaflet stiffness begins long before the clinical situa-tion mandates prosthetic replacement. Of the patientsreceiving echocardiographs in the Carpentier-Edwardsseries of Nakano and coworkers [16], 35% revealed sub-clinical prosthetic dysfunction, indicating that a hemody-namic impairment may cause damage for years.It was suggested that bioprostheses in the tricuspidposition have a lower rate of degeneration than in sys-temic circulation [17], but in our experience the rate of   Fig 4. Actuarial incidence of replacement of tricuspid prosthesis. Pericard   pericardial; prosth.   prosthesis; reop  reoperation; tot.  total. Table 5. Causes of Replacement of the Tricuspid Prosthesis TV ProsthesisIndication StructuralTotalProphylactic Degeneration Dehiscence Pannus Failure ThrombosisSmeloff-Cutter 0 0 0 1 0 0 1Pemco 0 0 0 0 1 0 1Lillehei 0 0 0 0 0 1 1Bjork delrin 0 0 0 0 0 1 1Bjork pirolitic 0 0 0 0 0 1 1Hancock 1 8 0 1 0 0 10Liotta 1 0 0 0 0 0 1Ionescu-Shiley 0 0 1 0 0 0 1Total 2 8 1 2 1 3 17 TV  tricuspid valve. S66  CARDIAC BIOPROSTHESES RIZZOLI ET AL Ann Thorac SurgTRICUSPID VALVE PROSTHESIS 1998;66:S62–7
Advertisement
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks