Impact of Prosthesis–Patient Size on Functional Recovery After Aortic Valve Replacement
http://www.100md.com
《循环学杂志》
the Departments of Cardiothoracic Anesthesia (C.G.K.) and Biostatistics (F.K.)
the Division of Anesthesia (F.G.E.)
the Department of Thoracic and Cardiovascular Surgery (F.D.L., E.H.B.)
The Cleveland Clinic Foundation, Cleveland, Ohio.
Abstract
Background— Prosthesis–patient size mismatch results when an implanted prosthetic aortic valve is of insufficient size for a patient’s body surface area. The relation between prosthesis–patient size and functional capacity and adverse postoperative outcome is inconsistent. Our objectives were to examine the impact of valve replacement, continuous prosthesis–patient size, and other factors on functional recovery after aortic valve replacement (AVR) with the Duke Activity Status Index (DASI).
Methods and Results— From June 15, 1995, through May 14, 1998, 1108 patients underwent AVR after completing a DASI survey. Of these, 1014 completed a postoperative DASI survey at an average of 8.3 months postoperatively. Logistic ordinal regression was used to examine the influence of demographic variables, comorbidities, baseline DASI scores, indexed valve orifice area, standardized orifice size, and postoperative morbid events on postoperative DASI. There was overall improvement in postoperative functional recovery reflected by median preoperative and postoperative DASI scores of 29 and 46, P<0.001, respectively. Neither indexed orifice area, P=0.94, nor standardized orifice size, P=0.96, was associated with functional recovery. Female sex, increasing age, elevated serum creatinine, increased central venous pressure, and red blood cell transfusion were factors associated with poor postoperative functional recovery.
Conclusions— A majority of patients report improvement in functional quality of life early after AVR. Similar functional recovery was demonstrated for patients along the full spectrum of valve sizes indexed to body size, even for values considered to represent severe mismatch for patient size. Factors other than prosthesis–patient size influence functional quality of life early after AVR.
Key Words: prosthesis ; stenosis ; surgery ; valves ; quality of life
Introduction
The term "prosthesis–patient mismatch" has been described as a prosthesis size in relation to body size that is less than that of a native aortic valve.1 Although residual gradients persist secondary to the nature of the structural support for prosthetic valves, patients with mismatch have been reported to exhibit continued degrees of left ventricular outflow tract obstruction, elevated transvalvular gradients, and less regression of left ventricular hypertrophy.2 The relation between prosthesis–patient size and adverse outcomes after aortic valve replacement (AVR) is inconsistent, and its impact on functional quality of life has not been adequately explored. Our objectives were to examine (1) the impact of valve replacement, (2) prosthesis–patient size expressed on a continuous scale, and (3) other variables on functional health-related quality of life after AVR with the Duke Activity Status Index (DASI; Figure 1).3 Our hypothesis was that smaller valve size relative to body size would be reflected by less improvement in functional quality of life postoperatively.
See p 3186
Methods
Patient Population and Data Collection
June 15, 1995, through May 14, 1998, 1108 patients underwent AVR at the Cleveland Clinic Foundation after completing a self-administered preoperative DASI survey. If a patient was unable to independently complete the questionnaire, a research assistant administered the survey by reading the exact wording of the survey questions. Characteristics of the patients are listed in Tables 1 through 5 . Perioperative variables were prospectively collected concurrently with patient care by dedicated individuals trained in database management and entered into the Cleveland Clinic Foundation Department of Cardiothoracic Anesthesia Registry. Additional institutional databases utilized for the analysis were the Cardiovascular Information Registry and the Department of Cardiology Echocardiographic database. The Social Security Death Index was accessed to determine death status during the follow-up interval. Institutional review board approval was obtained to perform research analyses with these databases.
Postoperative nominal 6- and 12-month follow-up DASI surveys were completed by means of telephone interviews. Ninety-four of the 1108 AVR patients did not complete a follow-up DASI survey. Among these, 21 had died within the 12-month follow-up interval and 73 patients failed to respond. The online-only Data Supplement depicts a comparison of responders and nonresponders. Nonresponders had similar baseline DASI, demographics, and preoperative comorbidities. Diabetes mellitus and peripheral vascular disease were more commonly represented in the nonresponders. Postoperative morbidities were also similar between the 2 groups, except that the nonresponders more frequently had total intubation times >72 hours.
Duke Activity Status Index
The DASI is a disease-specific functional quality-of-life questionnaire validated for patients with cardiovascular disease.3,4 The 12-item instrument (see Figure 1) measures activities of daily living such as household tasks, ambulation, personal care, sexual function, and recreational activities.3 Each question is weighted by the metabolic cost of each activity. The positive values for each component of the score are summed to form the DASI score.3 Possible scores range between 0 and 58.2. Higher scores represent better physical functioning.
Prosthesis–Patient Size
We examined the full spectrum of prosthesis–patient size values without dichotomizing these by arbitrarily predefining values for prosthesis–patient size mismatch. Prosthesis–patient size was normalized to patient size and expressed as internal orifice area indexed to body surface area and as a standardized orifice size (Z score) with the use of a regression equation formulated by Capps and colleagues.5 Indexed internal orifice area (in square millimeters) was calculated from the prosthesis internal orifice diameter and the patient’s body surface area (in square meters). Standardized orifice size expresses the number of standard deviations by which the internal orifice diameter deviates from mean normal aortic annular diameter predicted for the patient’s body surface area.5
Statistical Analyses
Preliminary Analysis
Summary statistics for each variable, including demographics, comorbid conditions, laboratory values, operative variables, and postoperative morbid outcome events, were calculated before modeling. Noninformative imputation was used for missing values. Transformations of continuous variables were introduced, including logistic, exponential, and inverse transforms. A comparison of 6- and 12-month DASI responses revealed few differences, with 61% of patients having similar DASI scores. Hence, only the first follow-up score obtained was used. Average interval from baseline to this follow-up was 8.3 months. Before modeling the data multivariably, univariable results were computed by comparing each variable by valve type with an ANOVA for normally distributed continuous variables, Kruskal-Wallis tests for nonnormally distributed continuous variables, and 2 and Fisher’s exact tests for categorical variables. Overall baseline and follow-up DASI score comparisons were made with the use of the Wilcoxon signed-rank test. Exploratory analyses were also conducted by scatterplots and histograms to reveal trends between DASI scores, age, and orifice size.
Modeling
Given the nature of the distribution of DASI scores, a logistic ordinal regression was used to model the multivariable effect of multiple variables on follow-up DASI score. Forming an ordinal scale addressed the problem of gaps in the distribution of DASI scores. To decide on an accurate grouping of the responses, several characteristics of the data were considered, including the natural grouping of the total scores, with cutpoints determined by deciles of the data and intercept estimates for each value treated as its own category. The resulting ordinal scale for the score was 0
All variables, including baseline DASI, were considered in the logistic ordinal regression modeling. To meet the logit-linear assumption of the ordinal regression model, age was transformed as the exponent of age in years/50. Backward selection with retention P<0.05 was used to construct a parsimonious model of the data.
To support the reliability of variable selection and identify the appropriate variable transformations, bootstrap bagging (aggregation) was used with an entry criterion of P<0.26 and a retention criterion of P<0.05 for a total of 200 samples of 1108 observations. After accounting for clustered variables, those variables or clusters identified 50% of the time or more were thought to be reliable and were placed in the ordinal model under the appropriate transformation. The proportional odds assumption was met; 2 test score P=0.13.
To account for confounding and patient-related factors that were part of the decision-making process of choosing a specific prosthesis, a balancing score was constructed for prosthesis–patient size. A balancing score is a type of multivariable statistical method that identifies individuals with similar predicted prosthesis–patient size. This balancing score was produced by linear regression analysis that included variables from Tables 1 through 5. The balancing score as well as the standardized orifice size and indexed orifice area were forced into the final ordinal regression model. All results were computed with SAS 8.2 software (SAS Institute).
Results
AVR and Functional Quality of Life
A majority of patients demonstrated improvement in functional quality of life after AVR, as reflected by the median preoperative and postoperative DASI scores of 29 and 46, P<0.001, respectively. Figure 3 displays baseline DASI scores and the change from baseline to first follow-up DASI scores for the patient population. Note the "ceiling effect"; ie, patients with high scores could remain the same or worsen. Those patients with low baseline scores improved or remained the same. Our patients with lower baseline DASI scores had a more noticeable improvement in scores compared with those with higher baseline scores, which, for the most part, stayed the same or decreased.
Impact of Prosthesis–Patient Size on Quality of Life
Impact of Other Variables on Quality of Life
Female sex, red blood cell transfusion, older age, higher preoperative serum creatinine, and increased postoperative central venous pressure were related to less favorable postoperative functional recovery (see Table 6). Figure 5 depicts the predicted probability for 2 follow-up DASI score groups for women and men by age from the multivariable ordinal regression model (Table 6). Values for other variables in the model included standardized orifice size of –0.6, median preoperative DASI of 31, preoperative creatinine of 1.3, central venous pressure of 12 mm Hg, and no blood transfusion. For every age group, women had a lower predicted probability than men of achieving the highest DASI functional category. A higher preoperative DASI score was associated with higher postoperative DASI scores (see Table 6).
Discussion
AVR and Functional Quality of Life
The purpose of AVR is to eliminate left ventricular outflow tract obstruction, facilitate regression of left ventricular hypertrophy, and, for patients with aortic stenosis, to relieve symptoms of heart failure, syncope, angina, and sudden death. Ultimately, improvements in a patient’s functional capacity and quality of life are desirable end points.
We report improvement in functional quality of life for a majority of patients after AVR. Other investigations of quality of life after AVR have also reported overall improvement in most quality-of-life domains.7–9 Additionally, a number of investigators have compared postoperative quality-of-life scores with established population norms and have reported similar or more favorable postoperative scores.7,10–12
Impact of Prosthesis–Patient Size on Quality of Life
Although the goal is to place an adequately sized aortic prosthesis for a patient’s size, it is not always possible, because of insufficient aortic annular dimensions.13,14 The prevalence of mismatch between prosthesis and patient size varies according to how the term is defined. A number of investigators predefine measures of prosthesis–patient size mismatch based on prior studies relating a specific cutoff value to increases in prosthesis pressure gradient.13,15 Others have calculated the ratio between the prosthetic valve effective orifice area and body surface area and have chosen the lowest decile to represent prosthesis–patient mismatch.2 Yet others do not dichotomize or predefine prosthesis–patient size mismatch; rather, they treat it as a continuous variable and examine outcomes.16 Measures of valve mismatch used in published investigations have included the calculated ratio between the prosthetic valve effective orifice area2,13 or internal orifice area16 and body surface area.
The impact of prosthesis–patient size mismatch on adverse outcomes after AVR is inconsistent. Patients defined as having prosthesis–patient size mismatch are reported to be at increased risk for short-2,15,17 and long-2 term mortality, deterioration in hemodynamic variables,13,17 exercise-induced increases in mean transvalvular pressure gradients approaching mild to moderate degrees of aortic stenosis,14 reduced exercise tolerance,18 and less extensive regression of left ventricular hypertrophy.19,20 Prosthesis–patient mismatch has been reported to be an independent predictor for postoperative congestive heart failure21 and has been associated with higher postoperative New York Heart Association (NYHA) functional class.17–19 Furthermore, those with mismatch reportedly have a higher occurrence of postoperative syncope, pulmonary edema, and angina.13
Moderate residual transvalvular pressure gradients have been associated with less regression of left ventricular hypertrophy and less improvement in left ventricular function.22 A lower percentage of left ventricular mass reduction has been reported to be among a number of variables independently predictive of heart failure and death postoperatively.23 Moreover, left ventricular hypertrophy has been implicated as a "causal link" between aortic stenosis and myocardial ischemia, ventricular arrhythmias associated with sudden death,20 reduced coronary reserve,24 and impaired left ventricular filling and contractility.24 A number of investigators have reported on the association between left ventricular hypertrophy and adverse outcomes.24–29
There are, however, a number of investigations that report similar survival,17,30,31 valve-related morbidity,30 NYHA functional class, and left ventricular mass index30 for those with and without prosthesis–patient size mismatch.
The discrepancy among investigations with regard to a number of outcomes may be related to the variable definition of prosthesis–patient mismatch. Interestingly, by slightly varying the predefined cutoff for the definition of mismatch, Ruel and colleagues21 reported different results for the same data set. When mismatch was defined as an indexed effective orifice area of 0.80 cm2/m2 or less, it was among a number of variables significantly related to postoperative heart failure after AVR. When mismatch was defined with a different cutoff point of 0.85 cm2/m2 or less, there was no significant association between mismatch and postoperative heart failure.
Goldsmith and colleagues32 were unable to report anything predictive of change in health 3 months after AVR when comparing mean preoperative baseline and 3-month follow-up responses to the Medical Outcomes Trust Short Form 36-item (SF-36) questionnaire. They suggested that patients with large valves had significant improvement in all life dimensions, whereas those with smaller valves were less likely to carry out daily social activities. The small sample size hindered their ability to examine whether this was due to patients’ small valve size or to older patients who more commonly had smaller implanted valve prostheses.
Although we hypothesized that smaller valve size indexed to body size would be related to lower postoperative DASI scores, we found that measures of indexed valve size were unrelated to postoperative functional recovery. Rather, other variables influenced functional quality of life after AVR.
Impact of Other Variables on Quality of Life
A number of preoperative and operative variables were related to functional recovery after AVR. Female sex was associated with poor functional recovery after adjusting for indexed orifice area measurements, standardized orifice size, baseline DASI scores, and a number of demographic variables, comorbidities, and postoperative morbid events. Although studies examining sex and quality-of-life outcomes after AVR are limited, a number of investigations have reported a reduced quality of life for women after coronary artery bypass grafting.33–35
Kirsch and colleagues36 reported that female sex was an independent predictor of impaired autonomy as well as late death for patients 80 years and older undergoing cardiac surgical procedures. They commented that women require smaller prostheses secondary to smaller aortic roots and that the smaller prostheses with more residual stenosis may be the factor that is related to poor late outcome.
We demonstrated that increasing age was associated with less functional recovery after AVR. Increasing age was associated with a lower likelihood of achieving higher postoperative DASI functional categories. Interpretation of the odds ratio for age in the model, 0.68 (confidence interval=0.58 to 0.78), is that the DASI score decreases at a faster than linear rate with advancing age. This is in contrast to a number of investigators who have reported improved quality of life for elderly patients undergoing cardiac surgical procedures when comparisons are made with the patients’ baseline values7,11,37–41 or with a general population norm.8,11
Olsson and colleagues37 reported similar improvements in quality of life 3 months after AVR for 32 octogenarians compared with 30 younger patients. Differences in quality of life between the older and younger groups and differences in intra-individual data were measured by examining postscore differences between the 2 groups and simple change scores.
Sedrakyan and colleagues7 reported no association between age and overall physical or mental health functioning 18 months after valve surgery, according to the mental and physical component summaries of the SF-36 survey. Multivariable regression techniques on change scores were used for the analysis. This lack of association between age and quality of life outcomes may be related, in part, to the statistical approach taken in the analysis of quality-of-life data. There is an increased risk for introducing bias when simple comparisons are made in postoperative follow-up scores or raw change scores, because differences in baseline values are not adequately controlled for. Furthermore, a regression on change score may additionally not control for differences in baseline values because of regression to the mean.38
Additionally, patients who are highly functional preoperatively are restricted by a ceiling effect, which limits their ability to achieve scores in a higher functional category postoperatively. Hence, these patients demonstrate less of a change in score when compared with those patients who have more baseline impairment, as reflected by lower quality-of-life scores. Sedrakyan and colleagues7 alluded to this concept for subgroups of patients with baseline scores nearly equivalent to population norms. These authors reported that patients aged 64 years or younger had significant improvements in all domains except general health and role emotional. On further examination, they reported that preoperative values for these domains were comparable to population norms, thereby "limiting possibility of recording higher absolute improvements." Again, for older patients in whom preoperative values were comparable to population norms, there was little or no improvement reported for those domains.7
Red blood cell transfusion, elevated preoperative creatinine, and increased postoperative central venous pressure were also associated with less functional recovery after AVR. The relation between these variables and quality of life after AVR has not been thoroughly explored. Specific preoperative valve pathophysiological features, such as aortic stenosis or regurgitation, were not significantly or reliably related to functional recovery postoperatively. Rather, age, which is a distinguishing factor for aortic stenosis versus regurgitation, was strongly associated with postoperative functional recovery.
The strength of this investigation is in the preoperative baseline and follow-up collection of quality-of-life data. A number of studies that have examined quality-of-life outcomes lack baseline surveys or have administered baseline surveys in the postoperative period, thereby potentially introducing bias, because only survivors of the operation would have completed a baseline survey.7,39,40 Furthermore, we did not approach the quality-of-life data with customary methods of analysis because of the shortcomings associated with these approaches. The use of an ordinal regression model to analyze the quality-of-life data added strength to this investigation.
Limitations of the Study
Limitations of the study include those factors that are related to an observational study design, in which unaccounted-for variables could influence the final results. Physical limitations such as degenerative joint disease may differentially impact the sexes and may be playing a role in the observed differences in DASI scores between women and men in our investigation. However, our investigation adjusted for preoperative baseline DASI scores in the modeling, which should adjust for preoperative disability from degenerative joint disease.
Conclusions
Overall functional quality of life is improved for patients undergoing AVR. Prosthesis–patient size does not influence functional recovery after AVR, even for those measures that would be considered severely mismatched for patient size. Preoperative variables such as baseline DASI scores, female sex, increasing age, higher preoperative serum creatinine, increased postoperative central venous pressure, and perioperative red blood cell transfusion exert considerable influence on postoperative functional recovery.
Acknowledgments
We thank Beth Lieber from the Department of Biostatistics for programming the data set. We also thank the physician assistants in the Department of Thoracic and Cardiovascular Surgery for assisting in the administration of the DASI instrument and the Department of Cardiothoracic Anesthesia Registry team for entering the information into the registry.
Footnotes
The online-only Data Supplement can be found with this article at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.104.505248/DC1.
References
Rahimtoola S. The problem of valve prosthesis–patient mismatch. Circulation. 1978; 58: 20–24.
Rao V, Jamieson E, Ivanov J, Armstrong S, David T. Prosthesis–patient mismatch affects survival after aortic valve replacement. Circulation. 2000; 102 (suppl III): III-5–III-9.
Hlatky M, Boineau R, Higginbotham M, Leek, Mark D, Califf R. A brief self-administered questionnaire to determine functional capacity (the Duke Activity Status Index) Am J Cardiol. 1989; 64: 651–654.
Nelson C, Herndon J, Mark D, Prior D, Califf R. Relation of clinical and angiographic factors to functional capacity as measured by the Duke Activity Status Index. Am J Cardiol. 1991; 68: 973–975.
Capps S, Elkins R, Fronk D. Body surface area as a predictor of aortic and pulmonary valve diameter. J Thorac Cardiovasc Surg. 2000; 119: 975–982.
Breiman L. Bagging and predicting. Machine Learning. 1996; 24: 123–140.
Sedrakyan A, Vaccarino V, Paltiel D, Elefteriades J, Mattera J, Roumanis S, Lin Z, Krumholz H. Age does not limit quality of life improvement in cardiac valve surgery. J Am Coll Cardiol. 2003; 42: 1208–1214.
Sundt T, Bailey M, Moon M, Mendeloff E, Huddleston C, Pasque M, Barner H, Gay W. Quality of life after aortic valve replacement at the age of >80 years. Circulation. 2000; 102 (suppl III): III-70–III-74.
Chocron S, Etievent J, Viel J, Dussaucy A, Clement F, Alwan K, Neidhardt M, Schipman N. Prospective study of quality of life before and after open heart operations. Ann Thorac Surg. 1996; 61: 153–157.
Perchinsky M, Henderson C, Jamieson W, Anderson W, Lamy A, Lowe N, de Guzman S. Quality of life in patients with bioprosthesis and mechanical prostheses: evaluation of cohorts of patients aged 51–65 years at implantation. Circulation. 1998; 98 (suppl II): II-81–II-86.
Fruitman D, MacDougall C, Ross D. Cardiac surgery in octogenarians: can elderly patients benefit; Quality of life after cardiac surgery. Ann Thorac Surg. 1999; 68: 2129–2135.
Pupello D, Bessone L, Lopez E, Brock J, Alkire M, Izzo E, Sanabria G, Sims D, Ebra G. Long-term results on the bioprosthesis in elderly patients: impact on quality of life. Ann Thorac Surg. 2001; 71: S244–S248.
Pibarot P, Honos G, Duranding L, Dumesnil J. The effect of prosthesis-patient mismatch on aortic bioprosthetic valve hemodynamic performance and patient clinical status. Can J Cardiol. 1996; 12: 379–387.
Pibarot P, Dumesnil J. Hemodynamic and clinical impact of prosthesis-patient mismatch in the aortic valve position and its prevention. J Am Coll Cardiol. 2000; 36: 1131–1141.
Blais C, Dumensil J, Baillot RM Simard S, Doyle D, Pibarot P. Impact of valve prosthesis–patient mismatch on short-term mortality after aortic valve replacement. Circulation. 2003; 108: 983–988.
Blackstone E, Cosgrove D, Jamieson W, Birkmeyer N, Lemmer J, Miller D, Butchart E, Rizzoli G, Yacoub M, Chai A. Prosthesis size and long-term survival after aortic valve replacement. J Thorac Cardiovasc Surgery. 2003; 126: 783–796.
Pibarot P, Dumesnil J, Lemieux M, Cartier P, Metras J, Durand L. Impact of prosthesis–patient mismatch on hemodynamic and symptomatic status, morbidity and mortality after aortic valve replacement with a bioprosthetic heart valve. J Heart Valve Dis. 1998; 7: 211–218.
de Carlo M, Milano A, Musumeci G, Tartarini G, Biadi O, Benedetti M, Bortolotti U. Cardiopulmonary exercise testing in patients with 21-mm St. Jude medical aortic prosthesis. J Heart Valve Dis. 1999; 8: 522–529.
Milano A, De Carlo M, Mecozzi G, D’Alfonso A, Scioti G, Nardi C, Bortolotti U. Clinical outcome in patients with 19-m and 21-mm St. Jude aortic prostheses: comparison at long-term follow-up. Ann Thorac Surg. 2002; 73: 37–43.
Gonzalez-Juanatey J, Garca-Acuna J, Fernandez M, Cendon A, Fuentes V, Garcia-Bengoechea J, de la Pena M. Influence of the size of aortic valve prostheses on hemodynamics and change in left ventricular mass: implications for the surgical management of aortic stenosis. J Thorac Cardiovasc Surgery. 1996; 112: 273–280.
Ruel M, Rubens F, Masters R, Pipe A, Bedard P, Hendry P, Khanh Lam B, Burwash I, Goldstein W, Brais M, Keon W, Mesana T. Late incidence and predictors of persistent or recurrent heart failure in patients with aortic prosthetic valves. J Thorac Cardiovasc Surg. 2004; 127: 149–159.
Jin X, Zhang Z, Gibson D, Yacoub M, Pepper J. Effects of valve substitute on changes in left ventricular function and hypertrophy after aortic valve replacement. Ann Thorac Surg. 1996; 62: 683–690.
Lessick J, Mutlak D, Markiewicz W, Reisner SA. Failure of left ventricular hypertrophy to regress after surgery for aortic stenosis. Echocardiography. 2002; 19: 359–366.
Messerli F, Ketelhut R. Left ventricular hypertrophy: an independent risk factor. J Cardiovasc Pharmacol. 1991; 17S: S59–S66.
Levy D, Garrison R, Savage D, Kannel W, Castelli W. Prognostic implications of echocardiographically determined left ventricular mass in the Framingham Heart Study. N Engl J Med. 1990; 322: 1561–1566.
Kannel WB. Left ventricular hypertrophy as a risk factor: the Framingham experience. J Hypertens Suppl. 1991; 9: S3–S8.
Levy D, Anderson K Savage D, Balkus S, Kannel W, Castelli W. Risk of ventricular arrhythmias in left ventricular hypertrophy: the Framingham Heart Study. Am J Cardiol. 1987; 60: 560–565.
Lonn E, Mathew J, Pogue J, Johnstone D, Danisa K, Baird B, Dagenais G, Sleight P, Yusuf S. Relationship of electrocardiographic left ventricular hypertrophy to mortality and cardiovascular morbidity in high-risk patients. J Cardiovasc Risk. 2003; 10: 420–428.
Haider A, Larson M, Benjamin E, Levy D. Increased left ventricular mass and hypertrophy are associated with increased risk for sudden death. J Am Coll Cardiol. 1998; 32: 1454–1459.
Hanayama N, Christakis G, Mallidi H, Joyner C, Fremes S, Morgan C, Mitoff P, Goldman B. Patient prosthesis mismatch is rare after aortic valve replacement: valve size may be irrelevant. Ann Thorac Surg. 2002; 73: 1822–1829.
Medalion B, Blackstone E, Lytle B, White J, Arnold J, Cosgrove D. Aortic valve replacement: is valve size important; J Thorac Cardiovasc Surg. 2000; 119: 963–974.
Goldsmith I, Lip G, Patel R. A prospective study of changes in quality of life of patients following mitral valve repair and replacement. Eur J Cardiothorac Surg. 2001; 20: 949–955.
Sjoland H, Wiklund I, Caidahl K, Haglid M, Westberg S, Herlitz J. Improvement in quality of life and exercise after coronary bypass surgery. Arch Intern Med. 1996; 156: 265–271.
Vaccarino V, Lin Z, Kasl S, Mattera J, Roumanis S, Abramson J, Krumholz H. Gender differences in recovery after coronary artery bypass surgery. J Am Coll Cardiol. 2003; 41: 307–314.
Koch C, Khandwala F, Cywinski J, Ishwaran H, Estafanous F, Loop F, Blackstone E. Health related quality of life following coronary artery bypass grafting: a gender analysis utilizing the Duke Activity Status Index. J Thorac Cardiovasc Surg. 2004; 128: 284–295.
Kirsch M, Guesnier L, LeBesnerais P, Hillion M, Debauches M, Seguin J, Loisance D. Cardiac operations in octogenarians: perioperative risk factors for death and impaired autonomy. Ann Thorac Surg. 1998; 66: 60–67.
Olsson M, Janfjal H, Orth-Gomer, Unden A, Rosenqvist M. Quality of life in octogenarians after valve replacement due to aortic stenosis: a prospective comparison with younger patients. Eur Heart J. 1996; 17: 583–589.
Vickers AJ, Altman DG. Statistics notes: analyzing controlled trials with baseline and follow up measurements. BMJ. 2001; 323: 1123–1124.
Immer F, Donati O, Wyss T, Immer-Bansi A, Schmidli Berdat P, Carrel T. Quality of life after mitral valve surgery: differences between reconstruction and replacement. J Heart Valve Dis. 2002; 12: 162–168.
Notzold A, Huppe M, Schmidtke C, Blomer P, Uhlig T, Sievers H. Quality of life in aortic valve replacement: pulmonary autografts versus mechanical prostheses. J Am Coll Cardiol. 2001; 37: 1963–1966.
Kumar P, Zehr K, Chang A, Cameron D, Baumgartner W. Quality of life in octogenarians after open heart surgery. Chest. 1995; 108: 919–926.(Colleen Gorman Koch, MD, )
the Division of Anesthesia (F.G.E.)
the Department of Thoracic and Cardiovascular Surgery (F.D.L., E.H.B.)
The Cleveland Clinic Foundation, Cleveland, Ohio.
Abstract
Background— Prosthesis–patient size mismatch results when an implanted prosthetic aortic valve is of insufficient size for a patient’s body surface area. The relation between prosthesis–patient size and functional capacity and adverse postoperative outcome is inconsistent. Our objectives were to examine the impact of valve replacement, continuous prosthesis–patient size, and other factors on functional recovery after aortic valve replacement (AVR) with the Duke Activity Status Index (DASI).
Methods and Results— From June 15, 1995, through May 14, 1998, 1108 patients underwent AVR after completing a DASI survey. Of these, 1014 completed a postoperative DASI survey at an average of 8.3 months postoperatively. Logistic ordinal regression was used to examine the influence of demographic variables, comorbidities, baseline DASI scores, indexed valve orifice area, standardized orifice size, and postoperative morbid events on postoperative DASI. There was overall improvement in postoperative functional recovery reflected by median preoperative and postoperative DASI scores of 29 and 46, P<0.001, respectively. Neither indexed orifice area, P=0.94, nor standardized orifice size, P=0.96, was associated with functional recovery. Female sex, increasing age, elevated serum creatinine, increased central venous pressure, and red blood cell transfusion were factors associated with poor postoperative functional recovery.
Conclusions— A majority of patients report improvement in functional quality of life early after AVR. Similar functional recovery was demonstrated for patients along the full spectrum of valve sizes indexed to body size, even for values considered to represent severe mismatch for patient size. Factors other than prosthesis–patient size influence functional quality of life early after AVR.
Key Words: prosthesis ; stenosis ; surgery ; valves ; quality of life
Introduction
The term "prosthesis–patient mismatch" has been described as a prosthesis size in relation to body size that is less than that of a native aortic valve.1 Although residual gradients persist secondary to the nature of the structural support for prosthetic valves, patients with mismatch have been reported to exhibit continued degrees of left ventricular outflow tract obstruction, elevated transvalvular gradients, and less regression of left ventricular hypertrophy.2 The relation between prosthesis–patient size and adverse outcomes after aortic valve replacement (AVR) is inconsistent, and its impact on functional quality of life has not been adequately explored. Our objectives were to examine (1) the impact of valve replacement, (2) prosthesis–patient size expressed on a continuous scale, and (3) other variables on functional health-related quality of life after AVR with the Duke Activity Status Index (DASI; Figure 1).3 Our hypothesis was that smaller valve size relative to body size would be reflected by less improvement in functional quality of life postoperatively.
See p 3186
Methods
Patient Population and Data Collection
June 15, 1995, through May 14, 1998, 1108 patients underwent AVR at the Cleveland Clinic Foundation after completing a self-administered preoperative DASI survey. If a patient was unable to independently complete the questionnaire, a research assistant administered the survey by reading the exact wording of the survey questions. Characteristics of the patients are listed in Tables 1 through 5 . Perioperative variables were prospectively collected concurrently with patient care by dedicated individuals trained in database management and entered into the Cleveland Clinic Foundation Department of Cardiothoracic Anesthesia Registry. Additional institutional databases utilized for the analysis were the Cardiovascular Information Registry and the Department of Cardiology Echocardiographic database. The Social Security Death Index was accessed to determine death status during the follow-up interval. Institutional review board approval was obtained to perform research analyses with these databases.
Postoperative nominal 6- and 12-month follow-up DASI surveys were completed by means of telephone interviews. Ninety-four of the 1108 AVR patients did not complete a follow-up DASI survey. Among these, 21 had died within the 12-month follow-up interval and 73 patients failed to respond. The online-only Data Supplement depicts a comparison of responders and nonresponders. Nonresponders had similar baseline DASI, demographics, and preoperative comorbidities. Diabetes mellitus and peripheral vascular disease were more commonly represented in the nonresponders. Postoperative morbidities were also similar between the 2 groups, except that the nonresponders more frequently had total intubation times >72 hours.
Duke Activity Status Index
The DASI is a disease-specific functional quality-of-life questionnaire validated for patients with cardiovascular disease.3,4 The 12-item instrument (see Figure 1) measures activities of daily living such as household tasks, ambulation, personal care, sexual function, and recreational activities.3 Each question is weighted by the metabolic cost of each activity. The positive values for each component of the score are summed to form the DASI score.3 Possible scores range between 0 and 58.2. Higher scores represent better physical functioning.
Prosthesis–Patient Size
We examined the full spectrum of prosthesis–patient size values without dichotomizing these by arbitrarily predefining values for prosthesis–patient size mismatch. Prosthesis–patient size was normalized to patient size and expressed as internal orifice area indexed to body surface area and as a standardized orifice size (Z score) with the use of a regression equation formulated by Capps and colleagues.5 Indexed internal orifice area (in square millimeters) was calculated from the prosthesis internal orifice diameter and the patient’s body surface area (in square meters). Standardized orifice size expresses the number of standard deviations by which the internal orifice diameter deviates from mean normal aortic annular diameter predicted for the patient’s body surface area.5
Statistical Analyses
Preliminary Analysis
Summary statistics for each variable, including demographics, comorbid conditions, laboratory values, operative variables, and postoperative morbid outcome events, were calculated before modeling. Noninformative imputation was used for missing values. Transformations of continuous variables were introduced, including logistic, exponential, and inverse transforms. A comparison of 6- and 12-month DASI responses revealed few differences, with 61% of patients having similar DASI scores. Hence, only the first follow-up score obtained was used. Average interval from baseline to this follow-up was 8.3 months. Before modeling the data multivariably, univariable results were computed by comparing each variable by valve type with an ANOVA for normally distributed continuous variables, Kruskal-Wallis tests for nonnormally distributed continuous variables, and 2 and Fisher’s exact tests for categorical variables. Overall baseline and follow-up DASI score comparisons were made with the use of the Wilcoxon signed-rank test. Exploratory analyses were also conducted by scatterplots and histograms to reveal trends between DASI scores, age, and orifice size.
Modeling
Given the nature of the distribution of DASI scores, a logistic ordinal regression was used to model the multivariable effect of multiple variables on follow-up DASI score. Forming an ordinal scale addressed the problem of gaps in the distribution of DASI scores. To decide on an accurate grouping of the responses, several characteristics of the data were considered, including the natural grouping of the total scores, with cutpoints determined by deciles of the data and intercept estimates for each value treated as its own category. The resulting ordinal scale for the score was 0
All variables, including baseline DASI, were considered in the logistic ordinal regression modeling. To meet the logit-linear assumption of the ordinal regression model, age was transformed as the exponent of age in years/50. Backward selection with retention P<0.05 was used to construct a parsimonious model of the data.
To support the reliability of variable selection and identify the appropriate variable transformations, bootstrap bagging (aggregation) was used with an entry criterion of P<0.26 and a retention criterion of P<0.05 for a total of 200 samples of 1108 observations. After accounting for clustered variables, those variables or clusters identified 50% of the time or more were thought to be reliable and were placed in the ordinal model under the appropriate transformation. The proportional odds assumption was met; 2 test score P=0.13.
To account for confounding and patient-related factors that were part of the decision-making process of choosing a specific prosthesis, a balancing score was constructed for prosthesis–patient size. A balancing score is a type of multivariable statistical method that identifies individuals with similar predicted prosthesis–patient size. This balancing score was produced by linear regression analysis that included variables from Tables 1 through 5. The balancing score as well as the standardized orifice size and indexed orifice area were forced into the final ordinal regression model. All results were computed with SAS 8.2 software (SAS Institute).
Results
AVR and Functional Quality of Life
A majority of patients demonstrated improvement in functional quality of life after AVR, as reflected by the median preoperative and postoperative DASI scores of 29 and 46, P<0.001, respectively. Figure 3 displays baseline DASI scores and the change from baseline to first follow-up DASI scores for the patient population. Note the "ceiling effect"; ie, patients with high scores could remain the same or worsen. Those patients with low baseline scores improved or remained the same. Our patients with lower baseline DASI scores had a more noticeable improvement in scores compared with those with higher baseline scores, which, for the most part, stayed the same or decreased.
Impact of Prosthesis–Patient Size on Quality of Life
Impact of Other Variables on Quality of Life
Female sex, red blood cell transfusion, older age, higher preoperative serum creatinine, and increased postoperative central venous pressure were related to less favorable postoperative functional recovery (see Table 6). Figure 5 depicts the predicted probability for 2 follow-up DASI score groups for women and men by age from the multivariable ordinal regression model (Table 6). Values for other variables in the model included standardized orifice size of –0.6, median preoperative DASI of 31, preoperative creatinine of 1.3, central venous pressure of 12 mm Hg, and no blood transfusion. For every age group, women had a lower predicted probability than men of achieving the highest DASI functional category. A higher preoperative DASI score was associated with higher postoperative DASI scores (see Table 6).
Discussion
AVR and Functional Quality of Life
The purpose of AVR is to eliminate left ventricular outflow tract obstruction, facilitate regression of left ventricular hypertrophy, and, for patients with aortic stenosis, to relieve symptoms of heart failure, syncope, angina, and sudden death. Ultimately, improvements in a patient’s functional capacity and quality of life are desirable end points.
We report improvement in functional quality of life for a majority of patients after AVR. Other investigations of quality of life after AVR have also reported overall improvement in most quality-of-life domains.7–9 Additionally, a number of investigators have compared postoperative quality-of-life scores with established population norms and have reported similar or more favorable postoperative scores.7,10–12
Impact of Prosthesis–Patient Size on Quality of Life
Although the goal is to place an adequately sized aortic prosthesis for a patient’s size, it is not always possible, because of insufficient aortic annular dimensions.13,14 The prevalence of mismatch between prosthesis and patient size varies according to how the term is defined. A number of investigators predefine measures of prosthesis–patient size mismatch based on prior studies relating a specific cutoff value to increases in prosthesis pressure gradient.13,15 Others have calculated the ratio between the prosthetic valve effective orifice area and body surface area and have chosen the lowest decile to represent prosthesis–patient mismatch.2 Yet others do not dichotomize or predefine prosthesis–patient size mismatch; rather, they treat it as a continuous variable and examine outcomes.16 Measures of valve mismatch used in published investigations have included the calculated ratio between the prosthetic valve effective orifice area2,13 or internal orifice area16 and body surface area.
The impact of prosthesis–patient size mismatch on adverse outcomes after AVR is inconsistent. Patients defined as having prosthesis–patient size mismatch are reported to be at increased risk for short-2,15,17 and long-2 term mortality, deterioration in hemodynamic variables,13,17 exercise-induced increases in mean transvalvular pressure gradients approaching mild to moderate degrees of aortic stenosis,14 reduced exercise tolerance,18 and less extensive regression of left ventricular hypertrophy.19,20 Prosthesis–patient mismatch has been reported to be an independent predictor for postoperative congestive heart failure21 and has been associated with higher postoperative New York Heart Association (NYHA) functional class.17–19 Furthermore, those with mismatch reportedly have a higher occurrence of postoperative syncope, pulmonary edema, and angina.13
Moderate residual transvalvular pressure gradients have been associated with less regression of left ventricular hypertrophy and less improvement in left ventricular function.22 A lower percentage of left ventricular mass reduction has been reported to be among a number of variables independently predictive of heart failure and death postoperatively.23 Moreover, left ventricular hypertrophy has been implicated as a "causal link" between aortic stenosis and myocardial ischemia, ventricular arrhythmias associated with sudden death,20 reduced coronary reserve,24 and impaired left ventricular filling and contractility.24 A number of investigators have reported on the association between left ventricular hypertrophy and adverse outcomes.24–29
There are, however, a number of investigations that report similar survival,17,30,31 valve-related morbidity,30 NYHA functional class, and left ventricular mass index30 for those with and without prosthesis–patient size mismatch.
The discrepancy among investigations with regard to a number of outcomes may be related to the variable definition of prosthesis–patient mismatch. Interestingly, by slightly varying the predefined cutoff for the definition of mismatch, Ruel and colleagues21 reported different results for the same data set. When mismatch was defined as an indexed effective orifice area of 0.80 cm2/m2 or less, it was among a number of variables significantly related to postoperative heart failure after AVR. When mismatch was defined with a different cutoff point of 0.85 cm2/m2 or less, there was no significant association between mismatch and postoperative heart failure.
Goldsmith and colleagues32 were unable to report anything predictive of change in health 3 months after AVR when comparing mean preoperative baseline and 3-month follow-up responses to the Medical Outcomes Trust Short Form 36-item (SF-36) questionnaire. They suggested that patients with large valves had significant improvement in all life dimensions, whereas those with smaller valves were less likely to carry out daily social activities. The small sample size hindered their ability to examine whether this was due to patients’ small valve size or to older patients who more commonly had smaller implanted valve prostheses.
Although we hypothesized that smaller valve size indexed to body size would be related to lower postoperative DASI scores, we found that measures of indexed valve size were unrelated to postoperative functional recovery. Rather, other variables influenced functional quality of life after AVR.
Impact of Other Variables on Quality of Life
A number of preoperative and operative variables were related to functional recovery after AVR. Female sex was associated with poor functional recovery after adjusting for indexed orifice area measurements, standardized orifice size, baseline DASI scores, and a number of demographic variables, comorbidities, and postoperative morbid events. Although studies examining sex and quality-of-life outcomes after AVR are limited, a number of investigations have reported a reduced quality of life for women after coronary artery bypass grafting.33–35
Kirsch and colleagues36 reported that female sex was an independent predictor of impaired autonomy as well as late death for patients 80 years and older undergoing cardiac surgical procedures. They commented that women require smaller prostheses secondary to smaller aortic roots and that the smaller prostheses with more residual stenosis may be the factor that is related to poor late outcome.
We demonstrated that increasing age was associated with less functional recovery after AVR. Increasing age was associated with a lower likelihood of achieving higher postoperative DASI functional categories. Interpretation of the odds ratio for age in the model, 0.68 (confidence interval=0.58 to 0.78), is that the DASI score decreases at a faster than linear rate with advancing age. This is in contrast to a number of investigators who have reported improved quality of life for elderly patients undergoing cardiac surgical procedures when comparisons are made with the patients’ baseline values7,11,37–41 or with a general population norm.8,11
Olsson and colleagues37 reported similar improvements in quality of life 3 months after AVR for 32 octogenarians compared with 30 younger patients. Differences in quality of life between the older and younger groups and differences in intra-individual data were measured by examining postscore differences between the 2 groups and simple change scores.
Sedrakyan and colleagues7 reported no association between age and overall physical or mental health functioning 18 months after valve surgery, according to the mental and physical component summaries of the SF-36 survey. Multivariable regression techniques on change scores were used for the analysis. This lack of association between age and quality of life outcomes may be related, in part, to the statistical approach taken in the analysis of quality-of-life data. There is an increased risk for introducing bias when simple comparisons are made in postoperative follow-up scores or raw change scores, because differences in baseline values are not adequately controlled for. Furthermore, a regression on change score may additionally not control for differences in baseline values because of regression to the mean.38
Additionally, patients who are highly functional preoperatively are restricted by a ceiling effect, which limits their ability to achieve scores in a higher functional category postoperatively. Hence, these patients demonstrate less of a change in score when compared with those patients who have more baseline impairment, as reflected by lower quality-of-life scores. Sedrakyan and colleagues7 alluded to this concept for subgroups of patients with baseline scores nearly equivalent to population norms. These authors reported that patients aged 64 years or younger had significant improvements in all domains except general health and role emotional. On further examination, they reported that preoperative values for these domains were comparable to population norms, thereby "limiting possibility of recording higher absolute improvements." Again, for older patients in whom preoperative values were comparable to population norms, there was little or no improvement reported for those domains.7
Red blood cell transfusion, elevated preoperative creatinine, and increased postoperative central venous pressure were also associated with less functional recovery after AVR. The relation between these variables and quality of life after AVR has not been thoroughly explored. Specific preoperative valve pathophysiological features, such as aortic stenosis or regurgitation, were not significantly or reliably related to functional recovery postoperatively. Rather, age, which is a distinguishing factor for aortic stenosis versus regurgitation, was strongly associated with postoperative functional recovery.
The strength of this investigation is in the preoperative baseline and follow-up collection of quality-of-life data. A number of studies that have examined quality-of-life outcomes lack baseline surveys or have administered baseline surveys in the postoperative period, thereby potentially introducing bias, because only survivors of the operation would have completed a baseline survey.7,39,40 Furthermore, we did not approach the quality-of-life data with customary methods of analysis because of the shortcomings associated with these approaches. The use of an ordinal regression model to analyze the quality-of-life data added strength to this investigation.
Limitations of the Study
Limitations of the study include those factors that are related to an observational study design, in which unaccounted-for variables could influence the final results. Physical limitations such as degenerative joint disease may differentially impact the sexes and may be playing a role in the observed differences in DASI scores between women and men in our investigation. However, our investigation adjusted for preoperative baseline DASI scores in the modeling, which should adjust for preoperative disability from degenerative joint disease.
Conclusions
Overall functional quality of life is improved for patients undergoing AVR. Prosthesis–patient size does not influence functional recovery after AVR, even for those measures that would be considered severely mismatched for patient size. Preoperative variables such as baseline DASI scores, female sex, increasing age, higher preoperative serum creatinine, increased postoperative central venous pressure, and perioperative red blood cell transfusion exert considerable influence on postoperative functional recovery.
Acknowledgments
We thank Beth Lieber from the Department of Biostatistics for programming the data set. We also thank the physician assistants in the Department of Thoracic and Cardiovascular Surgery for assisting in the administration of the DASI instrument and the Department of Cardiothoracic Anesthesia Registry team for entering the information into the registry.
Footnotes
The online-only Data Supplement can be found with this article at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.104.505248/DC1.
References
Rahimtoola S. The problem of valve prosthesis–patient mismatch. Circulation. 1978; 58: 20–24.
Rao V, Jamieson E, Ivanov J, Armstrong S, David T. Prosthesis–patient mismatch affects survival after aortic valve replacement. Circulation. 2000; 102 (suppl III): III-5–III-9.
Hlatky M, Boineau R, Higginbotham M, Leek, Mark D, Califf R. A brief self-administered questionnaire to determine functional capacity (the Duke Activity Status Index) Am J Cardiol. 1989; 64: 651–654.
Nelson C, Herndon J, Mark D, Prior D, Califf R. Relation of clinical and angiographic factors to functional capacity as measured by the Duke Activity Status Index. Am J Cardiol. 1991; 68: 973–975.
Capps S, Elkins R, Fronk D. Body surface area as a predictor of aortic and pulmonary valve diameter. J Thorac Cardiovasc Surg. 2000; 119: 975–982.
Breiman L. Bagging and predicting. Machine Learning. 1996; 24: 123–140.
Sedrakyan A, Vaccarino V, Paltiel D, Elefteriades J, Mattera J, Roumanis S, Lin Z, Krumholz H. Age does not limit quality of life improvement in cardiac valve surgery. J Am Coll Cardiol. 2003; 42: 1208–1214.
Sundt T, Bailey M, Moon M, Mendeloff E, Huddleston C, Pasque M, Barner H, Gay W. Quality of life after aortic valve replacement at the age of >80 years. Circulation. 2000; 102 (suppl III): III-70–III-74.
Chocron S, Etievent J, Viel J, Dussaucy A, Clement F, Alwan K, Neidhardt M, Schipman N. Prospective study of quality of life before and after open heart operations. Ann Thorac Surg. 1996; 61: 153–157.
Perchinsky M, Henderson C, Jamieson W, Anderson W, Lamy A, Lowe N, de Guzman S. Quality of life in patients with bioprosthesis and mechanical prostheses: evaluation of cohorts of patients aged 51–65 years at implantation. Circulation. 1998; 98 (suppl II): II-81–II-86.
Fruitman D, MacDougall C, Ross D. Cardiac surgery in octogenarians: can elderly patients benefit; Quality of life after cardiac surgery. Ann Thorac Surg. 1999; 68: 2129–2135.
Pupello D, Bessone L, Lopez E, Brock J, Alkire M, Izzo E, Sanabria G, Sims D, Ebra G. Long-term results on the bioprosthesis in elderly patients: impact on quality of life. Ann Thorac Surg. 2001; 71: S244–S248.
Pibarot P, Honos G, Duranding L, Dumesnil J. The effect of prosthesis-patient mismatch on aortic bioprosthetic valve hemodynamic performance and patient clinical status. Can J Cardiol. 1996; 12: 379–387.
Pibarot P, Dumesnil J. Hemodynamic and clinical impact of prosthesis-patient mismatch in the aortic valve position and its prevention. J Am Coll Cardiol. 2000; 36: 1131–1141.
Blais C, Dumensil J, Baillot RM Simard S, Doyle D, Pibarot P. Impact of valve prosthesis–patient mismatch on short-term mortality after aortic valve replacement. Circulation. 2003; 108: 983–988.
Blackstone E, Cosgrove D, Jamieson W, Birkmeyer N, Lemmer J, Miller D, Butchart E, Rizzoli G, Yacoub M, Chai A. Prosthesis size and long-term survival after aortic valve replacement. J Thorac Cardiovasc Surgery. 2003; 126: 783–796.
Pibarot P, Dumesnil J, Lemieux M, Cartier P, Metras J, Durand L. Impact of prosthesis–patient mismatch on hemodynamic and symptomatic status, morbidity and mortality after aortic valve replacement with a bioprosthetic heart valve. J Heart Valve Dis. 1998; 7: 211–218.
de Carlo M, Milano A, Musumeci G, Tartarini G, Biadi O, Benedetti M, Bortolotti U. Cardiopulmonary exercise testing in patients with 21-mm St. Jude medical aortic prosthesis. J Heart Valve Dis. 1999; 8: 522–529.
Milano A, De Carlo M, Mecozzi G, D’Alfonso A, Scioti G, Nardi C, Bortolotti U. Clinical outcome in patients with 19-m and 21-mm St. Jude aortic prostheses: comparison at long-term follow-up. Ann Thorac Surg. 2002; 73: 37–43.
Gonzalez-Juanatey J, Garca-Acuna J, Fernandez M, Cendon A, Fuentes V, Garcia-Bengoechea J, de la Pena M. Influence of the size of aortic valve prostheses on hemodynamics and change in left ventricular mass: implications for the surgical management of aortic stenosis. J Thorac Cardiovasc Surgery. 1996; 112: 273–280.
Ruel M, Rubens F, Masters R, Pipe A, Bedard P, Hendry P, Khanh Lam B, Burwash I, Goldstein W, Brais M, Keon W, Mesana T. Late incidence and predictors of persistent or recurrent heart failure in patients with aortic prosthetic valves. J Thorac Cardiovasc Surg. 2004; 127: 149–159.
Jin X, Zhang Z, Gibson D, Yacoub M, Pepper J. Effects of valve substitute on changes in left ventricular function and hypertrophy after aortic valve replacement. Ann Thorac Surg. 1996; 62: 683–690.
Lessick J, Mutlak D, Markiewicz W, Reisner SA. Failure of left ventricular hypertrophy to regress after surgery for aortic stenosis. Echocardiography. 2002; 19: 359–366.
Messerli F, Ketelhut R. Left ventricular hypertrophy: an independent risk factor. J Cardiovasc Pharmacol. 1991; 17S: S59–S66.
Levy D, Garrison R, Savage D, Kannel W, Castelli W. Prognostic implications of echocardiographically determined left ventricular mass in the Framingham Heart Study. N Engl J Med. 1990; 322: 1561–1566.
Kannel WB. Left ventricular hypertrophy as a risk factor: the Framingham experience. J Hypertens Suppl. 1991; 9: S3–S8.
Levy D, Anderson K Savage D, Balkus S, Kannel W, Castelli W. Risk of ventricular arrhythmias in left ventricular hypertrophy: the Framingham Heart Study. Am J Cardiol. 1987; 60: 560–565.
Lonn E, Mathew J, Pogue J, Johnstone D, Danisa K, Baird B, Dagenais G, Sleight P, Yusuf S. Relationship of electrocardiographic left ventricular hypertrophy to mortality and cardiovascular morbidity in high-risk patients. J Cardiovasc Risk. 2003; 10: 420–428.
Haider A, Larson M, Benjamin E, Levy D. Increased left ventricular mass and hypertrophy are associated with increased risk for sudden death. J Am Coll Cardiol. 1998; 32: 1454–1459.
Hanayama N, Christakis G, Mallidi H, Joyner C, Fremes S, Morgan C, Mitoff P, Goldman B. Patient prosthesis mismatch is rare after aortic valve replacement: valve size may be irrelevant. Ann Thorac Surg. 2002; 73: 1822–1829.
Medalion B, Blackstone E, Lytle B, White J, Arnold J, Cosgrove D. Aortic valve replacement: is valve size important; J Thorac Cardiovasc Surg. 2000; 119: 963–974.
Goldsmith I, Lip G, Patel R. A prospective study of changes in quality of life of patients following mitral valve repair and replacement. Eur J Cardiothorac Surg. 2001; 20: 949–955.
Sjoland H, Wiklund I, Caidahl K, Haglid M, Westberg S, Herlitz J. Improvement in quality of life and exercise after coronary bypass surgery. Arch Intern Med. 1996; 156: 265–271.
Vaccarino V, Lin Z, Kasl S, Mattera J, Roumanis S, Abramson J, Krumholz H. Gender differences in recovery after coronary artery bypass surgery. J Am Coll Cardiol. 2003; 41: 307–314.
Koch C, Khandwala F, Cywinski J, Ishwaran H, Estafanous F, Loop F, Blackstone E. Health related quality of life following coronary artery bypass grafting: a gender analysis utilizing the Duke Activity Status Index. J Thorac Cardiovasc Surg. 2004; 128: 284–295.
Kirsch M, Guesnier L, LeBesnerais P, Hillion M, Debauches M, Seguin J, Loisance D. Cardiac operations in octogenarians: perioperative risk factors for death and impaired autonomy. Ann Thorac Surg. 1998; 66: 60–67.
Olsson M, Janfjal H, Orth-Gomer, Unden A, Rosenqvist M. Quality of life in octogenarians after valve replacement due to aortic stenosis: a prospective comparison with younger patients. Eur Heart J. 1996; 17: 583–589.
Vickers AJ, Altman DG. Statistics notes: analyzing controlled trials with baseline and follow up measurements. BMJ. 2001; 323: 1123–1124.
Immer F, Donati O, Wyss T, Immer-Bansi A, Schmidli Berdat P, Carrel T. Quality of life after mitral valve surgery: differences between reconstruction and replacement. J Heart Valve Dis. 2002; 12: 162–168.
Notzold A, Huppe M, Schmidtke C, Blomer P, Uhlig T, Sievers H. Quality of life in aortic valve replacement: pulmonary autografts versus mechanical prostheses. J Am Coll Cardiol. 2001; 37: 1963–1966.
Kumar P, Zehr K, Chang A, Cameron D, Baumgartner W. Quality of life in octogenarians after open heart surgery. Chest. 1995; 108: 919–926.(Colleen Gorman Koch, MD, )