(Circulation. 1996;94:2883-2889.)
© 1996 American Heart Association, Inc.
Articles |
Mayo Clinic/St Mary's Hospital (R.J.R.), Rochester, Minn; CTRD Center at The University of Alabama at Birmingham (D.C.N., J.K.K., R.C.B.); Brigham and Women's Hospital, Boston, Mass (L.W.S.); Mid-America Heart Institute of St Luke's Hospital (C.B.P.), Kansas City, Mo; Baylor College of Medicine/The Methodist Hospital (J.B.Y.), Houston, Tex; St Louis (Mo) University (L.W.M.); Washington University School of Medicine (J.L.K.), St Louis, Mo; and Ohio State University (G.J.H.), Columbus.
Correspondence to R.J. Rodeheffer, MD, Cardiovascular Division, East-16B, Mayo Clinic, Rochester, MN 55905.
| Abstract |
|---|
|
|
|---|
Methods and Results At the time of this analysis, the Cardiac Transplant Research Database included all 2749 patients transplanted from January 1, 1990, to June 30, 1994, in the 25 participating transplant centers. During this 4.5-year period, the median waiting time for recipients who received a transplant increased from 2.7 to 3.5 months (P<.0001), and the proportion of recipients whose status was urgent at transplantation increased from 41% to 60% (P<.0001). Donor ischemic time increased from 150 to 166 minutes (P<.0001), and the proportion of donors requiring pressor support increased from 68% to 85% (P<.0001). Despite these changes in practice, the 1-year survival rate remained constant at 84% during this 4.5-year interval. There was no significant difference in 1-year survival rate between urgent status patients (83%) and nonurgent status patients (85%) (P=.08).
Conclusions The widening gap between the number of waiting recipients and the number of donors has resulted in a continuing trend toward transplanting urgent status recipients and to a liberalization of donor acceptance criteria. Despite these changes, posttransplant survival has remained constant.
Key Words: transplantation survival heart failure
| Introduction |
|---|
|
|
|---|
Because of these concerns, attempts have been made to predict changes in transplant practice, waiting time, and postoperative survival in the 1990s with the use of transplant survival data from the late 1980s.12 13 15 Such predictions have the potential to influence transplant practice and policy in the US and elsewhere. A possible weakness in these projections of future transplant practice and survival is the lack of up-to-date data on transplant outcomes. To provide current information on changes in North American cardiac transplantation, we analyzed trends in recipient and donor management in all 2749 transplants from January 1, 1990, to June 30, 1994, in the Cardiac Transplant Research Database (CTRD).
| Methods |
|---|
|
|
|---|
Patient Population
This study analyzes the complete primary cardiac transplant cohort of 2749 patients from the 25 participating CTRD centers from January 1, 1990, to June 30, 1994 (see "Appendix"). Patients were followed up through June 30, 1994, with a mean follow-up period of 25.3 months (range, 0 to 53.4 months).
The mean recipient age was 51.3 years; 79% were male, and 85% were white. The cause of heart disease was ischemic cardiomyopathy in 53%, dilated cardiomyopathy in 37%, congenital heart disease in 2%, and other causes in the remaining 8%. The mean donor age was 29.4 years (range, 5 to 72 years); 70% were male, and 82% were white. (See Tables 1 and 2![]()
.)
|
|
The severity of circulatory failure and the urgency for transplantation were determined by ascertainment of the patient's clinical status according to the United Network for Organ Sharing (UNOS) definitions that were applicable during the time of the present study. Status 1 patients were defined as those who required intravenous inotropic support or mechanical devices (intra-aortic balloon pump, total artificial heart, or ventricular assist device) to maintain cardiac output or who required ventilator support; these patients were classified as "urgent" status. All other patients were classified as status 2 patients or "nonurgent" status.
Data Collection
Data on all patients and donors were collected through coded event forms and forwarded to the CTRD center at the University of Alabama. The data management center performed computer data entry, data verification, and analysis.16 Data were reviewed by use of range checks and other statistical methods; missing or suspicious data entries were verified or corrected. Transplants were further verified by comparison with UNOS records.
Data Analysis
We examined changing characteristics among donors and recipients by looking at average trends among all institutions collectively. Attempts were also made to examine the variability among institutions in certain patient characteristics. A multivariable analysis in the hazard function domain generated risk factors for early and intermediate-term mortality over a 4.5-year period.17 Other analytical methods included the
2 test, one-way ANOVA, and Kaplan-Meier actuarial survival analysis. Statistical significance was accepted at the level of P<.05.
| Results |
|---|
|
|
|---|
9% during the 4.5-year study period (Table 1
Recipient Characteristics
The mean recipient age was 51.3 years and showed no change over time. The proportion of female recipients rose from 19% to 25% (P=.05). The proportion of black recipients remained unchanged at 10%.
The median waiting time for patients who received a transplant increased progressively from 2.7 months in 1990 to 3.5 months in 1994 (P<.0001). Since 1990, the likelihood of waiting 6 months for a heart transplant increased from 26% to 36%, the likelihood of waiting 1 year doubled from 11% to 20%, and the likelihood of waiting 2 years increased sixfold from 1% to 6%. These increases in waiting times represent "minimal" estimates in that only patients who survived to receive a transplant are included in this analysis.
During this period, the proportion of recipients who were urgent status at the initial listing increased from 24% to 32% (P=.005). Furthermore, the proportion of patients who were urgent status at the time of transplantation increased from 41% in 1990 to 60% in 1994 (P<.0001). (See Table 1
.)
Donor Characteristics
The mean donor age was 29.4 years and showed no time-dependent change. The gender distribution of donors also remained unchanged, with 70% being male. The total number of white donors remained nearly constant while the overall percentage of white donors gradually decreased from 86% to 79% (P<.0009). The proportion of black donors increased from 8% to 13% (P=.01), and Hispanic (plus other) donors increased proportionately from 6% to 8% (P=.04). (See Table 2
.)
Echocardiographic evaluation of donors occurred in 81% of donors in 1990, but this practice became nearly universal by 1994. The proportion of donors who received inotropic support progressively increased during the period of study from 68% in 1990 to 85% by 1994 (P<.0001). In addition, there was a significant increase in the use of donors who required moderate and high levels of inotropic support over this time period. (See Table 3
.)
|
An increase in median ischemic time occurred during the 4.5-year study period, rising from 150 minutes in 1990 to 166 minutes in 1994 (P<.0001). The proportion of donors with ischemic times >3 hours rose from 30% to 39% during the study period. Heterotopic heart transplantation is now very rare, decreasing from 1% in 1990 to 0% of transplanted hearts in 1994. A variety of cardioplegia solutions were used during the study period; the proportion of donors who received University of Wisconsin solution rose from 3% to 22% (P<.0001) and the proportion who used Collins solution fell from 10% to 2% (P<.001).
Donor-Recipient Matching
The matching of donor and recipient blood groups remained constant. However, gender match decreased from 71% to 65% (P=.02), and identical racial match decreased from 75% to 70% (P=.008) during the 4.5-year period of the study. Donor-recipient weight mismatch, defined as donor weight <80% of recipient weight, occurred in 15% of male recipients in 1990 and in 25% of male recipients in 1994 (P<.0001). When all recipients were analyzed (male and female), donor-recipient weight mismatch was less apparent because of the smaller size of the average female recipient. There was no significant change in the proportion of patients with baseline pulmonary vascular resistance >4.0 Wood units. (See Table 4
.)
|
Immunotherapy
The use of cytolytic induction therapy declined significantly; patients receiving induction with OKT3 antibody or anti-thymocyte globulin decreased from 41% in 1990 to 25% in 1994 (P<.0001). Preoperative cyclosporine was administered in >99% of patients in 1990 but in only 50% of patients in 1994 (P<.0001).
Variability Among Institutions During the Study
Several factors were examined with regard to variability among the 25 CTRD centers. Institutional differences in recipient and donor age were small, but notable variance was observed in individual institution mean waiting time (range, 3 to 12 months) and mean donor ischemic time (range, 120 to 200 minutes). Among the institutions, there were differences in the proportion of patients who were urgent status at transplant, ranging from 23% to 80%. Pulmonary vascular resistance at transplantation averaged 2.5 Wood units among all institutions, with the institutional mean ranging from 1.9 to 3.3 Wood units. (See Table 5
.)
|
Survival and Mortality Risk Factors
An analysis of overall survival rate by year transplanted showed no secular changes (P=.9); the 1-month survival rate was 93% to 95%, and the 1-year survival rate was 83% to 85% (Fig 1
). The risk factors for death were analyzed and are listed in Table 6
. When survival rate was adjusted for risk factors, there was no secular change in the adjusted 1-year rate. When "high-risk" transplants (recipient age
60 years, ischemic time >240 minutes, or donor age >50 years) were examined for survival, there was no change in survival rate by year of transplant (Fig 2
).
|
|
|
When overall survival was compared between the 1387 nonurgent status recipients and the 1358 urgent status recipients, there was no statistically significant difference in posttransplant survival. The 1-year survival rate was 85% for nonurgent status patients and 83% for urgent status patients (P=.08) (Fig 3
). In four patients, the urgency status at transplant was unknown.
|
| Discussion |
|---|
|
|
|---|
Transplant Activity
The number of transplants performed in the CTRD centers increased steadily from 590 per year in 1990 to
644 per year in 1994. This trend is consistent with total 1990 to 1993 US transplant activity reported by UNOS, ie, an increase in US heart transplants from 2108 in 1990 to 2299 in 199318 and an increase in the size of the US heart transplant waiting list from 1320 at the beginning of 1990 to 2935 in June 1994.19
Changes in Recipients
The present study shows no significant change in recipient age or race but a borderline significant increase in the proportion of female recipients. However, there was a marked increase in the proportion of recipients initially listed as urgent (status 1), rising from 24% to 32% from 1990 to 1994. Furthermore, those listed as urgent status at the time of transplantation increased even more sharply from 41% to 60% of recipients. This observation suggests that patients listed as nonurgent status deteriorate before surgery and shift to urgent status before being transplanted. These findings are consistent with the UNOS data suggesting an increase in waiting list patients requiring intensive care support from 41% in 1988 to 45% in 199120 and indicate that this secular trend has accelerated markedly since 1991. This pattern has been even more pronounced in reports from some individual institutions.21 22
An increase in the proportion of patients requiring intensive care support and who are classified as urgent status is likely related to the increase in recipient waiting time, because the longer a patient with severe circulatory failure waits, the more likely it is that the patient will experience clinical deterioration. Indeed, the median waiting time for transplanted patients in the CTRD centers rose from 2.7 months in 1990 to 3.5 months in 1994, and the percentage of patients with waiting times >1 year rose from 11% in 1990 to 20% in 1994.
Pretransplant Clinical Status and Postoperative Survival
The success of transplantation has resulted in an increase in the number of patients being put on waiting lists for the procedure. It is likely that patients are now listed who would have been considered too well, too sick, or too old for transplantation a decade ago. This has widened the gap between the numbers of donors and recipients and given rise to concern about how the limited number of donors may best be used to minimize death of proposed recipients while on the waiting list and optimize postoperative survival, ie, to maximize the survival of the total pool of potential transplant patients. To this end, patients who are likely to survive as well with current medical therapy as with transplant should not be prematurely transplanted, nor should patients be transplanted whose clinical status has deteriorated to the point at which the probability of postoperative survival is significantly diminished.
Much attention has been directed recently toward improving patient selection.12 13 15 Many patients with severe circulatory failure are now able to be treated medically with 1- and 2-year survival and quality of life comparable to that achieved by transplantation.15 23 Concern has been raised that growth in the size of the waiting list, particularly in the pool of urgent (status 1) patients, has resulted in a circumstance in which an increasing proportion of donor hearts are being allocated to urgent status candidates. Our data show that this trend has accelerated from 1990 to 1994.
Sophisticated statistical modeling techniques, with the use of estimates for waiting list size, urgent status listing, and survival with and without transplantation, have been applied to estimate survival for the total pool of potential transplant candidates. Outcome modeling, performed with the use of available postoperative survival data for urgent and nonurgent status patients, suggests that urgent status patients have a significantly poorer survival than nonurgent status patients.12 15 This has raised concerns that the direction of current US transplant practice toward transplantation of more urgent status patients will eventually result in a suboptimal allocation of donor hearts to a higher-risk recipient population and a poorer survival for the total pool of potential transplant patients.12 These analyses12 15 used an estimate of 1-month postoperative survival rate in urgent status patients of 86% and nonurgent status survival rate of 94%, based on data from the International Society for Heart and Lung Transplant Registry covering the period 1980 to 1990.24 In contrast, data from the present study representing the 1990 to 1994 era indicate that urgent status transplantation has a 1-month survival rate of 93%, comparable to that of nonurgent status patients at 94%. Although there was a trend toward slightly poorer 1-year survival for urgent status patients (85% for nonurgent status versus 83% for urgent status recipients, P=.08), the fact that the difference did not achieve statistical significance in a population of 2749 patients indicates that the impact of urgent status candidacy on transplantation outcome is very small at the present time. Although the limited effect of recipient status on outcome compared with previous eras could be the result of better postoperative management, it could also result from "status creep," ie, earlier advancement from nonurgent to urgent status, which could have occurred after the listing categories were simplified by UNOS from several levels to two active statuses in 1988. It is also possible that inotropic or mechanical support is being used earlier to maintain patients in optimal condition before transplantation and, thus, to improve postoperative outcome. These observations underscore the importance of collecting and using the highest-quality current data in describing current trends in transplant practice and in modeling future transplant activity and outcomes.
Changes in Donors
The present study shows no changes in the mean donor age of 29 years or in gender distribution from 1990 to 1994. There was an increase in the proportion of black donors from 8% to 13% of all donors, which suggests more effective recent donor recruitment among minority populations than was the case from 1988 to 1991, when the UNOS reported a small increase (from 8.0% to 9.9%) in donor recruitment from black patients.25
A gradual increase in donor ischemic time was evident in the present study, rising from 150 minutes in 1990 to 166 minutes in 1994. The proportion of donor hearts with ischemic time >3 hours rose from 30% to 39%. Although these ischemic times are still within the conventional 4-hour limit, the trend may reflect a growing willingness to accept the potential risk of longer ischemic times so that every available donor can be used. It may also be related to greater confidence in current cardioplegia solutions and the trend toward more widespread use of University of Wisconsin solution for cardiac protection. The steady increase in the proportion of donors who require inotropic support, from 68% to 85%, also suggests a trend toward the use of what in the past may have been considered "marginal" donors. This is underscored by the increase in the proportion of donors who require medium or high levels of inotropic support. Likewise, the increased willingness to use donors who are <80% of recipient weight (more evident in the male recipients, who have larger body size and who are more likely to face a size-mismatch problem), which increased from 15% to 25% of male recipients, suggests another trend toward pushing beyond previously accepted donor guidelines. Despite these extensions of conventional criteria for donor acceptability, the overall 1-year survival rate of transplanted patients has remained constant at
84% from 1990 to 1994.
Immunotherapy
Our findings documented for the first time recent changes in posttransplant immunosuppression practice, ie, marked decreases in the use of preoperative cyclosporine and the use of postoperative cytolytic induction therapy. Despite the simultaneous decrease in both forms of immune management, a change that might be expected to increase early postoperative rejection episodes, the survival rates have not changed.
Institutional Variability
There was notable variability among the 25 CTRD institutions in several areas of transplant practice. In particular, the average donor age ranged from as low as 26 years to as high as 34 years, and mean ischemic times varied from 120 minutes to 200 minutes among the different centers. Waiting time, which averaged 5.9 months, varied from as short as 3.0 months to as long as 12.0 months. Local geographic factors, such as population density, the size of regional waiting lists, and the proportion of urgent to nonurgent status patients, likely influence this institutional variability.
Study Limitations
Our data do not include all transplants performed in the United States. However, we believe that the transplants included were performed in a representative range of institutions and that trends observed in the CTRD likely reflect those in the region as a whole.
The data, while providing much improved and more current estimates of posttransplant survival, do not include information on survival from the time of wait listing for transplant and, therefore, are not sufficient to enable performance of up-to-date statistical modeling of transplant survival from the time of listing for transplant. The outcomes in this study are based on patients who actually received a transplant. Additional studies, some of which are now being conducted within other CTRD databases, will be necessary to characterize these aspects of current practice and to construct projections of future trends.26
Conclusions
This study provides current data indicating that significant secular changes are taking place in US cardiac transplant practice. As the waiting time has increased steadily beyond the donor supply, a larger proportion of patients are being transplanted as urgent status candidates. In addition, there have been increases in donor ischemic time, in the use of donors requiring inotropic support, and in the use of donors previously considered to be undersized. There has also been a trend toward less intense immunosuppression in the early postoperative period. Despite these changes, which might be expected to compromise posttransplant outcomes, the 1-year survival rate remained constant at
84% from 1990 through 1994.
| Appendix |
|---|
|
|
|---|
CTRD Center at The University of Alabama at Birmingham: D.C. Naftel, PhD, R.C. Bourge, MD, J.K. Kirklin, MD, D.C. McGiffin, MD, T. Wiess, MPH, MSPH, A. Crosswy, B. Austin, L. Early, P. Holmes, M. Veazey, P. Sims, K. Hubbard, J. Brush.
Abbott Northwestern Hospital/Minneapolis Heart Institute, Minneapolis, Minn: M.R. Pritzker, MD, K.D. Lake, PharmD, M. O'Kane, MSN, S. Chapman, PharmD, F. Hoffman, RN, N. Seimers, RN, C. Jorgensen, MD, W. Pedersen, MD, L. Joyce, MD, F. Eales, MD, R.W. Emery, MD, T. Von Reuden, MD, P. Bruhn, RN, M. King, MD, K. Arom, MD.
Albuquerque (NM) Presbyterian Hospital: K.J. Heilman III, MD, D. Pacheco, RN, C. Moore, RN, S. Levin, MSW, P. Blair, RN.
Brigham and Women's Hospital, Boston, Mass: L.W. Stevenson, MD, G.H. Mudge, MD, J. Jarcho, MD, P. Johnson, MD, E. Loh, MD.
Cleveland (Ohio) Clinic Foundation: R.E. Hobbs, MD, G. Rincon, MD, C. Bott-Silverman, MD, P. McCarthy, MD, R. Stewart, MD, L. Platt.
University of Florida, Shands Hospital, Gainesville: R.M. Mill, Jr, MD, S. Selman, RN, C. McGinn, RN, T. Walker, RN, J. Pooser.
Henry Ford Hospital, Detroit, Mich: T.B. Levine, MD, A.B. Levine, MD, B. Narins, RN.
Hershey Medical Center, Penn State University, Hershey, Pa: J. Boehmer, MD, P. Frazier, RN, P. Coe, RN.
Indiana University Medical Center, Indianapolis: J. O'Donnell, MD, A. Darroca, RN, L. Hiles, RN, R.L. Caldwell, MD, R.K. Darragh, MD, T. Flaspholer, RN.
Loyola University Medical Center, Maywood, Ill: M.R. Costanzo, MD,* M. Johnson, MD,* W. Kao, MD,* E. Winkel, MD, G.M. Mullen, MD, A. Heroux, MD* (*currently at Rush-Presbyterian Medical Center).
Mid-America Heart Institute of St Luke's Hospital, Kansas City, Mo: C.B. Porter, MD, A.M. Borkon, MD, D.R. Bresnahan, MD, R.E. Genton, MD, N.D. Long, RN, S.K. Rowe, MD, E.H. Russell, RN.
Medical College of Virginia, Richmond: D. Tolman, ME, H. Ibrahim, MD, A. Guerraty, MD, G. Sneed, RN, J. Hanrahan.
Methodist Hospital of Indianapolis (Ind): D.E. Pitts, MD, P. Kirlin, MD, H. Hollbrook, MD, L. Wallace, RN, S. Mitchell, RN, L. Kemp, RN.
Mayo Clinic/St Mary's Hospital, Rochester, Minn: R.J. Rodeheffer, MD, B.S. Edwards, MD, R.P. Frantz, MD, L.J. Olson, MD, R.C. Daly, MD, C.G.A. McGregor, MD.
Alton Ochsner Medical Institutions, New Orleans, La: H.O. Ventura, MD, F.W. Smart, MD, D.D. Stapleton, MD, C.H. Van Meter, Jr, MD, M.R. Mehra, MD, D. Dumas-Hicks, RN.
Ohio State University, Columbus: R.C. Starling, MD, G.J. Haas, MD, G. Whitby, RN.
St Louis (Mo) University: L.W. Miller, MD, S.H. Jennison, MD, N. Noedel, RN.
Tampa (Fla) General Hospital: L. Brauner, RN, G.B. Cintron, MD.
Baylor College of Medicine/The Methodist Hospital, Houston, Tex: J.B. Young, MD, J.A. Farmer, MD, B. Cocanougher, BSN, S. Lanthier, RN.
Temple University Hospital, Philadelphia, Pa: H.J. Eisen, MD, A.A. Bove, MD, PhD, V. Jeevanandam, MD, I.L. Pina, MD, K.B. Margulies, MD, J.B. McClurken, MD, H. Krystopa, RN, T. Regillo, RN.
University of Alabama at Birmingham: R.C. Bourge, MD, J.K. Kirklin, MD, D. McGiffin, MD, T. Noreuil, MD, C. White-Williams, RN, P. Blood, RN, J. Breland, RN, M. Hubbard, RN.
University of California, Los Angeles: J. Kobashigawa, MD, M. Hamilton, MD, A. Sabad, BA, G. Cogert.
University of Cincinnati (Ohio) Medical Center: G. Bhat, PhD, MD.
University of Minnesota, Minneapolis: S.H. Kubo, MD, E. Braunlin, MD, PhD, S. Diehl, RN, G.S. Francis, MD, A.J. Bank, MD, S.J. Shumway, MD, R.M. Bolman III, MD, S. Ormaza, RN, K. Monson, RN, R. Diekman, RN, M. O'Rourke, RN, K. Strasburg, RN.
University of Texas, Southwestern Medical Center/St Paul Medical Center, Dallas, Tex: C.W. Yancy, Jr, MD, P. Kaiser, RN, W.S. Ring, MD, N. Johnson, RN, MD, B.J. Balwin, MD, V.P.H. Horn.
Washington University School of Medicine, St Louis, Mo: J.L. Kenzora, MD, J.G. Rogers, MD, E.M. Geltman, MD, J. Zuksky, RN, S. Moorhead, RN, J. Vassolo, RN, C. Richardson, RN, C. Pasque, RN.
| Footnotes |
|---|
*See the "Appendix" for a listing of the Cardiac Transplant Research Database centers and personnel.
Received January 9, 1996; revision received June 24, 1996; accepted July 8, 1996.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
Y. S. Tjang, G. J.M.G. van der Heijden, G. Tenderich, D. E. Grobbee, and R. Korfer Survival analysis in heart transplantation: results from an analysis of 1290 cases in a single center Eur. J. Cardiothorac. Surg., May 1, 2008; 33(5): 856 - 861. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Lietz, R. John, D. M. Mancini, and N. M. Edwards Outcomes in cardiac transplant recipients using allografts from older donors versus mortality on the transplant waiting list: Implications for donor selection criteria J. Am. Coll. Cardiol., May 5, 2004; 43(9): 1553 - 1561. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. F. Lewis, S. W. Tsang, J. C. Fang, G. H. Mudge, J. A. Jarcho, C. M. Flavell, A. Nohria, M. M. Givertz, G. S. Couper, J. G. Byrne, et al. Frequency and impact of delayed decisions regarding heart transplantation on long-term outcomes in patients with advanced heart failure J. Am. Coll. Cardiol., March 3, 2004; 43(5): 794 - 802. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Scharf, T. Merz, W. Kiowski, E. Oechslin, C. Schalcher, and H. P. Brunner-La Rocca Noninvasive Assessment of Cardiac Pumping Capacity During Exercise Predicts Prognosis in Patients With Congestive Heart Failure Chest, October 1, 2002; 122(4): 1333 - 1339. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. L. Dickstein Anesthesia for Heart Transplantation Seminars in Cardiothoracic and Vascular Anesthesia, July 1, 1998; 2(2): 131 - 139. [Abstract] [PDF] |
||||
![]() |
C. Y. Wang, Y. Naka, H. Liao, M. C. Oz, T. A. Springer, J.-C. Gutierrez-Ramos, and D. J. Pinsky Cardiac Graft Intercellular Adhesion Molecule-1 (ICAM-1) and Interleukin-1 Expression Mediate Primary Isograft Failure and Induction of ICAM-1 in Organs Remote From the Site of Transplantation Circ. Res., April 20, 1998; 82(7): 762 - 772. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1996 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |