| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 1994;108:709-718
© 1994 Mosby, Inc.
SURGERY FOR ACQUIRED HEART DISEASE |
Portland, Ore., Vancouver, B.C., Canada, and Stanford, Calif.
Received for publication Jan. 7, 1994. Accepted for publication May 10, 1994. Address for reprints: Gary L. Grunkemeier, PhD, St. Vincent Heart Institute, 9155 SW Barnes, No. 230, Portland, OR 97225
Abstract
Actuarial analysis, using nonparametric (e.g., life table or Kaplan-Meier) or parametric (statistical modeling) methods, is used to describe and compare survival probabilities by allowing for partial survival times (censoring). Although devised to describe freedom from death, this method has been extended to nonfatal complications, such as freedom from tissue valve failure. However, the risk described for nonfatal events is that which a patient would experience provided he were immortal. And patients with valve disease have a relatively high risk of dying, generating the question: "What is the chance the valve will fail before the patient dies? " To answer this more practical (for individual patient management and population resource allocation) question requires an estimate of what we call actual failure, that is, the percentage of patients whose valve will actually fail before they die. This risk is less than the risk which the usual actuarial curve describes. This difference increases with patient age, because older patients have a lower risk of tissue failure and a higher risk of death than younger patients. (J THORACCARDIOVASCSURG1994;108:709-18)
Actuarial analysis has been used for the evaluation of heart valve results for almost 30 years 
1 and is considered the standard for reporting time-related results. 2 We use this word to refer to the usual analysis of single end point, grouped interval data by the life table method 3 or individual patient data by the Kaplan-Meier product-limit method. 4 The actuarial event-free (survival) curve for a nonfatal event, such as structural valve deterioration (SVD) of porcine valves, estimates the event-free probability for a population in which death has been eliminated. This overestimates the percentage of valves that will actually fail, because many patients die before the valve fails. What is of greater interest for advising patients, or for expense planning for a population, is the risk of SVD occurring before death. There is no consensus on the name for this risk in the statistical literature; cumulative incidence is commonly used, but we also refer to this as actual risk.
We 5 previously used SVD results based on a visual fit of points from published curves to illustrate this concept. In the present report we use individual porcine valve follow-up data from two centers to fit models for actual and actuarial SVD; to incorporate the risk factors age, gender, and position; and to compare these parametric models with nonparametric fits.
PATIENTS AND METHODS
Patients
A total of 4910 Hancock and Carpentier-Edwards porcine valves (Hancock valves, Medtronic, Inc., Minneapolis, Minn.; Carpentier-Edwards valves, Baxter Healthcare Corp., Edwards Div., Santa Ana, Calif.) from two centers, including all operative survivors of isolated (nonmultiple) porcine valve replacement during 1975 to 1988, were available for analysis (Table I). These valves combined for 29,610 patient-years of follow-up, with a maximum of 15.0 years (Table II). Follow-up completeness, defined as a known date of death or explantation, or contact from 1988 through 1990, was 94%. Among these isolated valves, there were 516 cases of SVD (68 fatal) and 1349 deaths not related to SVD (Table II). In addition, 164 patients with double valve replacement (Table III) were used to compare their observed SVD experience with that predicted from the risk of their two separate valves.
|
|
|
Actuarial SVD. Nonparametric actuarial SVD-free curves were constructed by the Kaplan-Meier method. 4 For parametric analysis, we fit several probability functions commonly used in reliability analysis (including exponential, Weibull, Rayleigh, gamma, normal, and lognormal), evaluating the fits according to values of the likelihood function and visual comparison of fitted and empirical event-free and cumulative hazard function curves. The Weibull distribution, which is widely used to describe wear-out or fatigue failure data, 6 was chosen to model actuarial SVD.
Risk factors considered for the regression model were patient age and gender; valve position, size, and model; and year of implantation (several interaction terms were also considered during the model fitting). The maximum likelihood method was used to simultaneously estimate the shape parameter of the Weibull distribution and the coefficients of the risk factors that determine its scale parameter. 7
The parametric event-free percentage for a subgroup at a given postoperative time was computed by exponentiating the negative of the cumulative hazard, which is computed as the integral of the mean hazard, based on the regression model, up until that time. The mean hazard at each point is taken as the mean for all valves still potentially at risk (at that postoperative time), rather than the mean of all valves entering the series. 8
Double valve replacement data were not used to estimate the regression models. To see if the risk for an individual valve is changed in the double valve replacement environment, we computed the individual risk for each patient having double valve replacement based on the Weibull regression model for each valve separately, assuming the two valves would fail independently (i.e., added the individual risks, which is equivalent to multiplying the individual SVD-free curves), and compared this predicted risk with the nonparametric actuarial SVD-free curve for the patients having double valve replacement.
Actuarial patient survival. Nonparametric actuarial survival curves were constructed by the Kaplan-Meier method. 4 Parametric estimation of late survival used a regression model from a previous study, 9 based on the Gompertz distribution, which is widely used for survival, especially in older populations. 10 SVD-related deaths were excluded (censored) in the survival analysis, to estimate patient survival in the absence of valve failure. Less than 5% of late deaths were associated with SVD (see 
Table II).
Actual SVD. Nonparametric estimation of actual SVD involves a summation, 11,12 and the parametric model involves a counterpart to this sum (i.e., mathematical integration of the models for actuarial SVD and patient survival). 13 Methods for obtaining standard error and confidence interval estimates for these functions have been described for nonparametric 11,12 and parametric 13 estimates but are not implemented here. Our purpose is to describe this concept rather than make statistical comparisons; because of the large sample size, however, confidence intervals would be rather narrow.
RESULTS
Nonparametric actuarial SVD-free curves (Fig. 1, top) and cumulative incidence curves (Fig. 1, bottom) were both remarkably similar for the two centers. At 14 years, the actuarial SVD-free probabilities are 50% to 58%, whereas the cumulative incidence of SVD is only about 30%.
|
Table IV) because it is known beforehand, whereas size is not, and position has more often been found to influence SVD by other studies.
|
Table IV), and it was similar for various subsets. The Weibull model approximated the observed data when stratified by age group (Fig. 2) and position (Fig. 3).
|
|
|
|
|
|
|
Actuarial risk of death
If a series is followed up until all patients have died, then the percentage who have died at any given postoperative year can be given directly. However, in the usual ongoing series this simple percentage would underestimate the probability of death, because many patients are still alive at the time of reporting. Actuarial methods precisely accommodate the partial follow-up times of these patients, referred to as censored observations. If patients who are censored proceed on with the same risk of death as those who were not censored, then the actuarial method provides estimates of percent dead (or alive) at each year as if censoring were removed, that is, all patients had been followed up until death.
Actuarial risk of nonfatal events
The actuarial method is also used for nonfatal events, such as SVD, by including death as an additional censoring mechanism (Appendix A). This assumes that those who die are only temporarily suspended from the calculation and will continue on to experience the same SVD risk as their fellows. Thus, provided this assumption is true, this method yields estimates of percent SVD at each year as if death never occurs. This may be regarded as the intrinsic risk of the nonfatal event, but this risk does not have a direct interpretation for a patient.
Actual risk of nonfatal events
Of more direct clinical relevance is the percent experiencing SVD, given that death does intervene. This statistic is more useful for describing the outcome for an individual patient and for anticipating the future flow of returning patients for health care planning purposes. 13 Increased patient age has a doublyimportant effect on actual SVD, decreasing the actuarial (intrinsic) risk of the event (see Fig. 5) while increasing the risk of the competing event death (see Fig. 2).
Features of parametric models
We have used a parametric approach, based on separate models for (the actuarial risk of) SVD and death, to estimate actual SVD. Parametric estimation results in an equation that describes the entire experience over time, given a relatively few number of parameters; usually a constant (intercept), a shaping parameter, and some coefficients for the risk factors that constitute the scaling factor of the chosen statistical distribution (Appendix B). Parametric models provide smooth estimates that more realistically portray the continuity of nature, rather than the irregularities of one particular data set. 17 Using such models, one can compute the entire event-free curve for a given set of risk factors, even beyond the maximum observation time, and compute summary statistics such as mean and median survival times. This enables us to extrapolate the actual SVD curves beyond the 15 years of observation, to the point where they level off and thus estimate the percent of patients who will ever actually experience SVD (before death), as in Fig. 7. Of course the results from such extrapolations must be interpreted cautiously because they depend on assumptions regarding the distribution and risk factors of the model.
Risk factors for SVD
The regression models for actuarial SVD and death in this study were based on a limited number of risk factors. Our risk factors for SVD agree with the findings of others. Jones and associates 18 performed a Cox regression of the effect of age on SVD which, according to their graphs, agrees remarkably well with the results of the present study. Other studies have shown the continuous effect of age 19-21 and the effect of valve position 19-22 on the actuarial risk of SVD.
However, these models could be expanded by incorporating other potential risk factors for SVD (activity level, renal and hepatic function, anticoagulation, etc.) and late death (valve lesion, functional class, concomitant coronary bypass, etc.). Another set of factors associated with SVD could be porcine factors. Unlike mechanical valves, in which units of the same model may be considered almost identical, tissue valves possess inherent biologic variation because of donor and perhaps manufacturing factors. When tissue valves of the same type are tested in vitro under identical conditions, there is a wide variation in times to failure, 23,24 and in 24 patients who received sequential porcine valves, the second seemed to fare better than the first. 25
CONCLUSION
This article has two themes. A minor theme is that the Weibull distribution seems to fit the actuarial SVD curves for porcine valves, thus allowing (cautious) extrapolation beyond observed experience. The major theme is that the usual actuarial curves do not provide the most meaningful percentages of SVD, as far as patient prognosis is concerned. The first finding (combined with a previously developed model for patient survival) allows us to use parametric methods for the second; that is, to project (extrapolate) the percentage of patients who will ever experience SVD. One could use the nomogram in Fig. 8 to determine lower age bounds for the use of porcine valves, depending on the eventual risk of replacement one wished to tolerate. For example, a risk of 20% of ever experiencing SVD (lowest horizontal grid line) corresponds to a minimum age of 70 to 73 years; a risk of 40% corresponds to ages 59 to 63 years.
The present study used series that include the earliest use of first-generation porcine valves. With current application, the overall results can be expected to be better because of refinements in the newer generations of porcine and pericardial valves and better refined criteria for selecting which patients receive a bioprosthesis.
We thank Eva T. Germann, Florence Chan, and Kathy Moore for data acquisition and preparation and the reviewers of this paper for many helpful suggestions. We especially appreciate the insightful counsel of Eugene H. Blackstone, MD, with whom one of the authors has had many discussions concerning the subject of this paper, which resulted in great improvement in its contents and presentation.
Appendix: APPENDIXES
Appendix A: Basic survivorship functions
Single end point.
When time-related events are being evaluated statistically, a usual construct is the random variable T, which is the time to the event or end point. For example, in a clinical series, Tdeath is the time (from operation) to death. It can be characterized by any of a number of equivalent mathematical functions. The most fundamental is the cumulative distribution function:
F(t) = prob(Tdeath
t)
the probability that death occurs before time t. In clinical description, its complement, the familiar survival function, is more often used:
S(t) = 1–F(t) = prob (Tdeath> t)
the probability of survival to time t. These two functions add to unity for every time t.
There are other useful representations for computing and representing risk, such as the density function f(t) = F'(t), which is the mathematical derivative of F(t) with respect to time, and the hazard function h(t) = f(t)/S(t), which measures the instantaneous risk given survival to time t. But the survival and distribution functions are both expressed as cumulative probabilities and are easier to interpret, to appreciate the risk of an outcome, such as the chance of living until (or dying before) 5 years, 10 years, and so on.
Multiple end points—actuarial.
When the lifetime of a prosthetic valve is being considered, two (or more) risks compete as end points: failure of the valve (our primary concern here) and death of the patient, either of which will terminate its functioning. These events are associated with two times: TSVD and Tdeath.
The usual actuarial curve is derived with the assumption that death and SVD are statistically independent events. The distribution function is as follows:
FSVD(t) = prob(TSVD t and Tdeath
= infinite),
the probability the valve has failed by time t, with death removed as a risk. The survival function can be written as follows:
SSVD(t) = prob(TSVD > t and Tdeath = infinite)
that is, the probability the valve has not failed by time t, with death removed as a risk. The assumption of independence is not true in this situation, because those at higher risk for death are at lower risk for SVD.
Multiple endpoints—actual.
When the presence of death is acknowledged, there are two "distribution" functions:
prob(Tdeath < t and Tdeath < TSVD)
that is, the probability the patient died before t and before the valve (would have) failed. This might be called a success of valve selection;
prob(TSVD
t and TSVD
< Tdeath)
that is, the probability the valve failed before t and (obviously) before the patient died. This might be called a failure of valve selection. It measures the probability of SVD in the situation where death is acting. It is not a true distribution function because it is always less than one. In the statistical literature, the terms absolute risk, 13 crudeincidence, 11,16 cause-specific failure, 12 and cumulative incidence 12,14-16 have been used to describe functions, probabilities, and curves associated with this concept. The term actual risk is one we have introduced to refer to this concept, in phonetic contradistinction to the term actuarial. To emphasize that it is not an accepted statistical term, we have used parentheses or italics throughout the text.
The survival function which, with the two cumulative incidence functions described earlier adds to unity at every time t, is as follows:
Sboth(t) = prob(TSVD > t and Tdeath > t)
that is, the probability the valve has not failed and the patient has not died (the patient is alive with a functioning valve).
Appendix B: Parametric and nonparametric formulas
In this section, bold symbols (x,y) are used to indicate one or more risk factors. A general formula for a parametric event-free percentage at time t for a patient with risk factor(s) x is the exponential function of the negative of the cumulative hazard:
S(t|x) = exp (–R(x) · H(t))
The cumulative hazard above is itself the product of two functions: a common shaping function H(t), from a basic family of distributions, and a patient-specific scaling function R(x), which incorporates risk factor(s) into the model. For technical reasons, R(x) is usually an exponential function of the risk factors. For example, with two factors x1 = AGE and x2 = MITRAL:
R(x) = R(x1, x2) = exp (B0 + B1 · AGE + B2 · MITRAL)
where AGE is in years (or a transformation thereof) and MITRAL equals 1 for yes and 0 for no. The coefficients (B's) are estimated from the data.
Actuarial SVD.
The Weibull distribution used to model SVD has a cumulative hazard of the form:
(R · t)A=RA · tA
where the superscript indicates exponentiation. A is the shape parameter (aging factor) and R is the scale parameter. In the regression model, RA is replaced by R(x), a function of the risk factors x, and the actuarial SVD-free function becomes:
SSVD (t|x) = exp (–R(x) · tA)
Patient survival.
The Gompertz distribution used for late death has a cumulative hazard of the form:
R · (exp (A · t) –1)
where A is a shape parameter and R is a scale parameter. In the regression model, R is replaced by a function R(y) of the risk factors, and the patient survival function is as follows:
Sdead (t|y) = exp (–R(y) · (exp (A · t)–1))
Actual SVD.
The nonparametric, cumulative incidence function for SVD (see Fig. 6) at postimplantation time T is a sum, for all failure times ti < T:
where SBOTH (ti) is the nonparametric (Kaplan-Meier) estimate of freedom from both SVD and death at time ti, and di and ni are the numbers of SVD events and remaining valves at time ti, respectively. This summand is the product of the percent of (the original number of) valves still functioning at time ti and the percent failing at that time and is thus an estimate of the percent of the original valves failing at that time. Summing this over all times up to T provides an estimate of the percent of valves (actually) experiencing SVD by time T.
The parametric expression for the actual SVD percentage at T years (Figs. 6 to 8) for a patient with risk factors x for SVD and y for death is as follows:
where hSVD represents the instantaneous hazard function for SVD. This is the parametric counterpart to the previous formula, with summation replaced by integration and the overall survival function written as the product of the two survival functions for SVD and death. (There is a small omission here, because a third possible end point, nonfatal non-SVD valve explants, has been ignored; if incorporated, it would slightly lower the estimate of actual SVD.) Because the expressions inside the integral sign can be computed out to any time T, we can let T approach infinity (20 or so years, really) to estimate the percentage of patients who will ever experience SVD.
Footnotes
From St. Vincent Heart Institute, Sisters of Providence Health System, Portland, Ore., the Division of Cardiovascular and Thoracic Surgery, University of British Columbia, Vancouver, B.C., Canada, and Department of Cardiothoracic Surgery, Stanford University School of Medicine, Stanford, Calif. 
References
This article has been cited by other articles:
|
|
 |
|
  W. R. E. Jamieson, R. Koerfer, C. A. Yankah, A. Zittermann, R. I. Hayden, H. Ling, R. Hetzer, and W. B. Dolman Mitroflow aortic pericardial bioprosthesis -- clinical performance Eur. J. Cardiothorac. Surg., November 1, 2009; 36(5): 818 - 824. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A Kulik, B-K Lam, F D Rubens, P J Hendry, R G Masters, W Goldstein, P Bedard, T G Mesana, and M Ruel Gender differences in the long-term outcomes after valve replacement surgery Heart, February 1, 2009; 95(4): 318 - 326. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Lorusso, S. Gelsomino, G. De Cicco, C. Beghi, C. Russo, M. De Bonis, A. Colli, and A. Sala Mitral valve surgery in emergency for severe acute regurgitation: analysis of postoperative results from a multicentre study Eur. J. Cardiothorac. Surg., April 1, 2008; 33(4): 573 - 582. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. N. Anderson Comment on the "actual" problem editorial J. Thorac. Cardiovasc. Surg., February 1, 2008; 135(2): 457 - 459. [Full Text] [PDF]
|
|
|
|
 |
|
 R. F. Padera Jr. and F. J. Schoen Pathology of Cardiac Surgery Card. Surg. Adult, January 1, 2008; 3(2008): 111 - 178. [Full Text]
|
|
|
|
|
|
R. Jin and G. L. Grunkemeier Statistical Treatment of Surgical Outcome Data Card. Surg. Adult, January 1, 2008; 3(2008): 247 - 258. [Full Text]
|
|
|
|
|
|
N. D. Desai and G. T. Christakis Bioprosthetic Aortic Valve Replacement: Stented Pericardial and Porcine Valves Card. Surg. Adult, January 1, 2008; 3(2008): 857 - 894. [Full Text]
|
|
|
|
|
|
T. Gudbjartsson, T. Absi, and S. Aranki Mitral Valve Replacement Card. Surg. Adult, January 1, 2008; 3(2008): 1031 - 1068. [Full Text]
|
|
|
|
|
|
M. W.A. van Geldorp, W.R. E. Jamieson, A. P. Kappetein, J. P.A. Puvimanasinghe, M. J.C. Eijkemans, G. L. Grunkemeier, J. J.M. Takkenberg, and A. J.J.C. Bogers Usefulness of microsimulation to translate valve performance into patient outcome: Patient prognosis after aortic valve replacement with the Carpentier Edwards supra-annular valve J. Thorac. Cardiovasc. Surg., September 1, 2007; 134(3): 702 - 709. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Hayashida, K. Kanda, H. Yaku, J. Ando, and Y. Nakayama Development of an in vivo tissue-engineered, autologous heart valve (the biovalve): Preparation of a prototype model J. Thorac. Cardiovasc. Surg., July 1, 2007; 134(1): 152 - 159. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. L. Grunkemeier, R. Jin, M. J.C. Eijkemans, and J. J.M. Takkenberg Actual and Actuarial Probabilities of Competing Risks: Apples and Lemons Ann. Thorac. Surg., May 1, 2007; 83(5): 1586 - 1592. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Authors/Task Force Members, A. Vahanian, H. Baumgartner, J. Bax, E. Butchart, R. Dion, G. Filippatos, F. Flachskampf, R. Hall, B. Iung, et al. Guidelines on the management of valvular heart disease: The Task Force on the Management of Valvular Heart Disease of the European Society of Cardiology Eur. Heart J., January 26, 2007; (2007) ehl428v1. [Full Text] [PDF]
|
|
|
|
 |
|
 W. E. Jamieson, E. Germann, M. R Aupart, P. H Neville, M. A Marchand, and G. J Fradet 15-Year Comparison of Supra-Annular Porcine and PERIMOUNT Aortic Bioprostheses Asian Cardiovasc Thorac Ann, June 1, 2006; 14(3): 200 - 205. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Chan, W.R.E. Jamieson, E. Germann, F. Chan, R.T. Miyagishima, L.H. Burr, M.T. Janusz, H. Ling, and G.J. Fradet Performance of bioprostheses and mechanical prostheses assessed by composites of valve-related complications to 15 years after aortic valve replacement J. Thorac. Cardiovasc. Surg., June 1, 2006; 131(6): 1267 - 1273. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Pitfalls and limitations in measuring and interpreting the outcomes of mitral valve repair. J. Thorac. Cardiovasc. Surg., March 1, 2006; 131(3): 523 - 529.
|
|
|
|
|
|
E. Bodnar and E. H. Blackstone An "actual" problem: Another issue of apples and oranges J. Thorac. Cardiovasc. Surg., January 1, 2006; 131(1): 1 - 3. [Full Text] [PDF]
|
|
|
|
|
|
C. W. Akins, A. D. Hilgenberg, G. J. Vlahakes, J. C. Madsen, T. E. MacGillivray, G. M. LaMuraglia, and R. P. Cambria Late Results of Combined Carotid and Coronary Surgery Using Actual Versus Actuarial Methodology Ann. Thorac. Surg., December 1, 2005; 80(6): 2091 - 2097. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W.R.E. Jamieson, O. von Lipinski, R.T. Miyagishima, L.H. Burr, M.T. Janusz, H. Ling, G.J. Fradet, F. Chan, and E. Germann Performance of bioprostheses and mechanical prostheses assessed by composites of valve-related complications to 15 years after mitral valve replacement J. Thorac. Cardiovasc. Surg., June 1, 2005; 129(6): 1301 - 1308. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. J. Schoen and R. J. Levy Calcification of Tissue Heart Valve Substitutes: Progress Toward Understanding and Prevention Ann. Thorac. Surg., March 1, 2005; 79(3): 1072 - 1080. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Ruel, A. Kulik, F. D. Rubens, P. Bedard, R. G. Masters, A. L. Pipe, and T. G. Mesana Late incidence and determinants of reoperation in patients with prosthetic heart valves Eur. J. Cardiothorac. Surg., March 1, 2004; 25(3): 364 - 370. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Kaempchen, T. Guenther, M. Toschke, G.L. Grunkemeier, M. Wottke, and R. Lange Assessing the benefit of biological valve prostheses: cumulative incidence (actual) vs. Kaplan-Meier (actuarial) analysis Eur. J. Cardiothorac. Surg., May 1, 2003; 23(5): 710 - 714. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. J.M. Takkenberg, M. J.C. Eijkemans, L. A. van Herwerden, E. W. Steyerberg, M. M. Lane, R. C. Elkins, J.D. F. Habbema, and A. J.J.C. Bogers Prognosis after aortic root replacement with cryopreserved allografts in adults Ann. Thorac. Surg., May 1, 2003; 75(5): 1482 - 1489. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. J. Schoen and R. F. Padera Jr. Cardiac Surgical Pathology Card. Surg. Adult, January 1, 2003; 2(2003): 119 - 185. [Full Text]
|
|
|
|
|
|
G. Grunkemeier and R. Jin Statistical Treatment of Outcome Data Card. Surg. Adult, January 1, 2003; 2(2003): 225 - 231. [Full Text]
|
|
|
|
|
|
N. D. Desai and G. T. Christakis Stented Mechanical/Bioprosthetic Aortic Valve Replacement Card. Surg. Adult, January 1, 2003; 2(2003): 825 - 856. [Full Text]
|
|
|
|
|
|
T. Gudbjartsson, S. Aranki, and L. H. Cohn Mechanical/Bioprosthetic Mitral Valve Replacement Card. Surg. Adult, January 1, 2003; 2(2003): 951 - 986. [Full Text]
|
|
|
|
 |
|
 J. M. G. Paez, E. J. Herrero, A. Rocha, M. Martin-Maestro, J. L. Castillo-Olivares, I. Millani, A. C. Sanmartin, and A. Cordon The Telescoping Suture - Part II: A Novel Method to Improve the Mechanical Behavior of a New Biomaterial: Ostrich Pericardium J Biomater Appl, October 1, 2002; 17(2): 105 - 123. [Abstract] [PDF]
|
|
|
|
|
|
G. L. Grunkemeier and Y. Wu Interpretation of nonfatal events after cardiac surgery: Actual versus actuarial reporting J. Thorac. Cardiovasc. Surg., August 1, 2001; 122(2): 216 - 219. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. S. Khan, A. Trento, M. DeRobertis, R. M. Kass, M. Sandhu, L. S. C. Czer, C. Blanche, S. Raissi, G. P. Fontana, W. Cheng, et al. Twenty-year comparison of tissue and mechanical valve replacement J. Thorac. Cardiovasc. Surg., August 1, 2001; 122(2): 257 - 269. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. G. Paez, A. Carrera, E. J. Herrero, I. Millan, A. Rocha, A. Cordon, N. Sainz, J. Mendez, and J. L. Castillo-Olivares Influence of the Selection of the Suture Material on the Mechanical Behavior of a Biomaterial to be Employed in the Construction of Implants. Part 2: Porcine Pericardium J Biomater Appl, July 1, 2001; 16(1): 68 - 90. [Abstract] [PDF]
|
|
|
|
|
|
W.R. E. Jamieson, M. T. Janusz, L. H. Burr, H. Ling, R. T. Miyagishima, and E. Germann Carpentier-Edwards supraannular porcine bioprosthesis: second-generation prosthesis in aortic valve replacement Ann. Thorac. Surg., May 1, 2001; 71 (2007): S224 - S227. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Marchand, M. R. Aupart, R. Norton, I. R.A. Goldsmith, L. C. Pelletier, M. Pellerin, T. Dubiel, W. J. Daenen, P. Herijgers, F. P. Casselman, et al. Fifteen-year experience with the mitral Carpentier-Edwards PERIMOUNT pericardial bioprosthesis Ann. Thorac. Surg., May 1, 2001; 71 (2007): S236 - S239. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. A. Grossi, A. LaPietra, G. H. Ribakove, J. Delianides, R. Esposito, A. T. Culliford, C. C. Derivaux, R. M. Applebaum, I. Kronzon, B. M. Steinberg, et al. Minimally invasive versus sternotomy approaches for mitral reconstruction: Comparison of intermediate-term results J. Thorac. Cardiovasc. Surg., April 1, 2001; 121(4): 708 - 713. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. R. Moon, D. C. Miller, K. A. Moore, P. E. Oyer, R. S. Mitchell, R. C. Robbins, E. B. Stinson, N. E. Shumway, and B. A. Reitz Treatment of endocarditis with valve replacement: the question of tissue versus mechanical prosthesis Ann. Thorac. Surg., April 1, 2001; 71(4): 1164 - 1171. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. P. A. Puvimanasinghe, E. W. Steyerberg, J. J. M. Takkenberg, M. J. C. Eijkemans, L. A van Herwerden, A. J. J. C. Bogers, and J. D. F. Habbema Prognosis After Aortic Valve Replacement With a Bioprosthesis : Predictions Based on Meta-Analysis and Microsimulation Circulation, March 20, 2001; 103(11): 1535 - 1541. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. E. David, J. Ivanov, S. Armstrong, C. M. Feindel, and G. Cohen Late results of heart valve replacement with the Hancock II bioprosthesis J. Thorac. Cardiovasc. Surg., February 1, 2001; 121(2): 0268 - 278. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. J.O. Birkmeyer, J. D. Birkmeyer, A. N.A. Tosteson, G. L. Grunkemeier, C. A.S. Marrin, and G. T. O'Connor Prosthetic valve type for patients undergoing aortic valve replacement: a decision analysis Ann. Thorac. Surg., December 1, 2000; 70(6): 1946 - 1952. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. A. Grossi, P. K. Zakow, M. Sussman, A. C. Galloway, J. Delianides, G. Baumann, and S. B. Colvin Late results of mitral valve reconstruction in the elderly Ann. Thorac. Surg., October 1, 2000; 70(4): 1224 - 1226. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. H. Blackstone and B. W. Lytle COMPETING RISKS AFTER CORONARY BYPASS SURGERY: THE INFLUENCE OF DEATH ON REINTERVENTION J. Thorac. Cardiovasc. Surg., June 1, 2000; 119(6): 1221 - 1232. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. R. Eric Jamieson, M. A. Marchand, C. L. Pelletier, R. Norton, M. Pellerin, T. W. Dubiel, M. R. Aupart, W. J. Daenen, M. P. Holden, T. E. David, et al. STRUCTURAL VALVE DETERIORATION IN MITRAL REPLACEMENT SURGERY: COMPARISON OF CARPENTIER-EDWARDS SUPRA-ANNULAR PORCINE AND PERIMOUNT PERICARDIAL BIOPROSTHESES J. Thorac. Cardiovasc. Surg., August 1, 1999; 118(2): 297 - 305. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. S. Peterseim, Y.-Y. Cen, S. Cheruvu, K. Landolfo, T. M. Bashore, J. E. Lowe, W. G. Wolfe, and D. D. Glower LONG-TERM OUTCOME AFTER BIOLOGIC VERSUS MECHANICAL AORTIC VALVE REPLACEMENT IN 841 PATIENTS J. Thorac. Cardiovasc. Surg., May 1, 1999; 117(5): 890 - 897. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. G. Paez and E. Jorge-Herrero Assessment of Pericardium in Cardiac Bioprostheses: A Review J Biomater Appl, April 1, 1999; 13(4): 351 - 388. [Abstract] [PDF]
|
|
|
|
 |
|
 K. M. Kim, G. A. Herrera, and H. D. Battarbee Role of Glutaraldehyde in Calcification of Porcine Aortic Valve Fibroblasts Am. J. Pathol., March 1, 1999; 154(3): 843 - 852. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W.R. E. Jamieson, L. H. Burr, M. T. Janusz, A. I. Munro, R. I. Hayden, R. T. Miyagishima, H. Ling, G. J. Fradet, S. V. Lichtenstein, and K. M. Stewart Carpentier-Edwards standard and supraannular porcine bioprostheses: comparison of technology Ann. Thorac. Surg., January 1, 1999; 67(1): 10 - 17. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. C. McGiffin, A. J. Galbraith, M. F. O'Brien, G. J. McLachlan, D. C. Naftel, P. Adams, S. Reddy, and L. Early AN ANALYSIS OF VALVE RE-REPLACEMENT AFTER AORTIC VALVE REPLACEMENT WITH BIOLOGIC DEVICES J. Thorac. Cardiovasc. Surg., February 1, 1997; 113(2): 311 - 318. [Abstract] [Full Text]
|
|
|
|
|
|
J. I. Fann, D. C. Miller, KathleenA. Moore, R. S. Mitchell, P. E. Oyer, E. B. Stinson, R. C. Robbins, B. A. Reitz, and N. E. Shumway Twenty-Year Clinical Experience With Porcine Bioprostheses Ann. Thorac. Surg., November 1, 1996; 62(5): 1301 - 1311. [Abstract] [Full Text]
|
|
|
|
|
|
L. H. Edmunds Jr, R. E. Clark, L. H. Cohn, G. L. Grunkemeier, D. C. Miller, and R. D. Weisel Guidelines for Reporting Morbidity and Mortality After Cardiac Valvular Operations Ann. Thorac. Surg., September 1, 1996; 62(3): 932 - 935. [Abstract] [Full Text]
|
|
|
|
|
|
L. H. Edmunds Jr., R. E. Clark, L. H. Cohn, G. L. Grunkemeier, D. C. Miller, and R. D. Weisel GUIDELINES FOR REPORTING MORBIDITY AND MORTALITY AFTER CARDIAC VALVULAR OPERATIONS J. Thorac. Cardiovasc. Surg., September 1, 1996; 112(3): 708 - 711. [Full Text]
|
|
|
|
 |
|
 W. Vongpatanasin, L. D. Hillis, and R. A. Lange Prosthetic Heart Valves N. Engl. J. Med., August 8, 1996; 335(6): 407 - 416. [Full Text] [PDF]
|
|
|
|
|
|
A. K. Agnihotri, D. C. McGiffin, A. J. Galbraith, and M. F. O'Brien THE PREVALENCE OF INFECTIVE ENDOCARDITIS AFTER AORTIC VALVE REPLACEMENT J. Thorac. Cardiovasc. Surg., December 1, 1995; 110(6): 1708 - 1724. [Abstract] [Full Text]
|
|
|
|
|
|
D. M. Cosgrove, B. W. Lytle, P. C. Taylor, M. T. Camacho, R. W. Stewart, P. M. McCarthy, D. P. Miller, M. R. Piedmonte, and F. D. Loop The Carpentier-Edwards pericardial aortic valve: Ten-year results J. Thorac. Cardiovasc. Surg., September 1, 1995; 110(3): 651 - 662. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
