| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1995;60:284-290
© 1995 The Society of Thoracic Surgeons
Division of Cardiothoracic Surgery, Department of Surgery, Section of Surgical Sciences, Pennsylvania State University College of Medicine, University Hospital and Children's Hospital, Hershey, Pennsylvania
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentReferences |
|---|
Methods. The use of ventricular assistance devices (VADs) at Pennsylvania State University as a bridge to transplantation was reviewed. Records of 43 patients listed as status I from July 1991 to July 1994 were compared.
Results. This analysis demonstrated that for all patients treated with the intent to transplant, those who were bridged with a VAD exhibited a trend toward an improved transplantation rate (92% versus 68%) and a significantly greater rate of discharge from the hospital (92% versus 55.4%; p = 0.023) than the medically managed patients. Although overall charges and costs were higher in VAD-supported patients, this was related to significantly longer pretransplantation hospitalization. When normalized to daily costs and charges, this discrepancy in expenses was eliminated.
Conclusions. The superior rate of discharge at equitable daily costs and charges for the VAD patients draws continued enthusiasm toward use of these devices as a bridge to transplantation. Furthermore, development of outpatient care for VAD-supported patients and continued advances in the use of these devices may further reduce the cost of managing these critically ill patients.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
Despite nearly uniform acceptance of heart transplantation as the standard of care for suitable patients with end-stage heart failure [1], there has not been concomitant acceptance of mechanical ventricular assistance as a temporary means of supporting the failing patient. With increasing concerns about the escalating costs of medical care and the realities of medical economics, both research and application of this technology have come under great scrutiny [2]. Previous studies have cited the perceived increased costs associated with circulatory support before heart transplantation [3–5]. The interpretation of these studies, however, may be misleading for varied reasons. Many of these reviews have compared varied populations of patients. The inclusion of all patients during a period in which the technology for these devices has changed significantly [6–9] and failure to normalize costs and charges to the length of hospitalization associated with cardiac transplantation are examples of this problem. These issues require further investigation to ensure that accurate information is presented to those groups responsible for dictating decisions regarding the use of mechanical circulatory support, as well as agencies responsible for the continued funding of ongoing research toward the advancement of these devices.
Ventricular assistance as bridge to transplantation has been used at the Pennsylvania State University since 1985 [10, 11]. Over the last 5 years, implementation of mechanical assistance has been subjected to stricter and more uniform patient inclusion criteria before implantation of a device. Outcome results at our institution have remained consistent over this period and continue to mirror multicenter registry results [12].
Patients who undergo implantation of an assist device are classified as status I by United Network for Organ Sharing criteria. This classification is reserved for those patients who (1) have undergone implantation of a total artificial heart; (2) have undergone implantation of a right, left (LVAD), or biventricular assist device; (3) are being supported by an intraaortic balloon counterpulsation pump; (4) require mechanical ventilatory support; or (5) are both hospitalized in an intensive care unit and receiving inotropic support to sustain cardiac output [13]. This set of patients is considered to be the most critically ill and is by definition hospitalized secondary to need for extraordinary support of a failing heart. Thus, comparison of patients who have undergone implantation of a mechanical assist pump appears to be more accurate when limited to those patients who require chronic inotropic support or placement of an intraaortic balloon pump within a hospitalized setting.
The status I patients at the Pennsylvania State University over the last 4 years were chosen as a suitable population for comparison. A review of the local database resulted in patients being classified in one of two major groups. Patients either were mechanically assisted as a bridge to transplantation or were treated medically with inotropic agents or an intraaortic balloon pump with an intent to treat with orthotopic transplantation. A comparison between these two groups was undertaken with regard to a complement of variables including (1) total length of hospitalization, (2) duration of pretransplantation and posttransplantation hospitalization, (3) outcome statistics, and (4) costs and charges that were further normalized to length of stay to determine daily figures. These results were checked for significance and present a means for comparing these different approaches to the management of the patient with a chronically failing heart.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
). During this same period 15 patients underwent implantation of a Pierce-Donachy LVAD. All patients from this list who were classified by United Network for Organ Sharing criteria as status I at time of transplantation or death were entered into this study. These criteria excluded 3 patients who underwent implantation of an LVAD for postcardiotomy shock syndrome or acute myocardial infarction without intent to transplant. An additional 22 patients were excluded from inclusion to the study as 16 patients were classified as status II at time of death or transplantation and an additional 6 patients were inactivated from the transplant list before death.
|
The medical records were reviewed for each patient. In addition, the financial records for each patient were obtained from the Department of Clinical Cost Accounting. This allowed the total costs and charges accrued during hospitalization by each status I patient to be identified.
All patients who underwent transplantation have undergone complete follow-up evaluation at the University Hospital. The mean follow-up for group I was 11.2 months, and ranged from 2 to 32 months. Group II patients were followed up for an average of 12.4 months with a range of 2 to 35 months.
Admission Profile
The total length of the admission associated with status I classification was determined for each patient. Additionally, this period was subdivided to determine the duration of (1) time preceding implantation of an assist device; (2) interval before receiving a transplant, and (3) period from transplantation until discharge from the hospital. These intervals were quantified for each patient, and mean values were calculated for each group.
The transplantation and discharge rates were determined as a function of the total population for each group. Both transplantation and discharge rates were compared between groups to allow assessment of the differences in outcome when compared by mode of support before transplantation. Thus this constituted a comparison between two groups of patients treated with intent to transplant.
Cost Comparison
A summation of all patient charges and hospital costs during the admission with status I classification was undertaken for each patient. These values subsequently were adjusted to account for the significant discrepancy in length of admission for the two groups. Figures for each patient were normalized by factoring each as a function of the total number of days of hospitalization. Thus a charge/day and cost/day was calculated for each patient, and mean values for each group were compared for significant difference.
Statistical Analysis
All data were compiled with a personal-computer based software package (Minitab; Minitab, Inc, State College, PA). A two-sample t test and confidence interval was used for comparison of independent population means. Additionally, a 
2 analysis was used for discrete variables. Actuarial survival statistics were determined by Kaplan-Meier methods, and interval survival data were compared by log rank analysis. A confidence interval level of 95% was considered to be statistically significant (p < 0.05).
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
reflect that the age, sex, pretransplantation diagnosis, blood types, and average total waiting time for transplantation were similar between the two groups.
|
Evaluation of the length of time between transplantation and discharge is demonstrated in Table 2. The patients in group I spent an average of 17.8 days convalescing in the hospital before discharge. This was not significantly different from the average of 22.2 days of hospitalization after transplantation for the patients in group II (p = 0.33).
|
reflect that 92% of the patients in group I were maintained by mechanical assistance until a suitable donor organ was available. This rate proved better than group II, where only 68% of the patients survived to transplantation; however, this comparison failed to reach significance. Further analysis demonstrated that all the patient from group I who received transplants were discharged from the hospital; however, only 55.4% of the original patients in group II survived to discharge. This comparison reflects that the patients in group I had a significantly improved discharge rate.
|
. The patients in group I had a mean cost and charge of $186,131 and $302,048. These figures were significantly greater than those accrued by group II, which totaled $100,115 and $165,219 (p = 0.0001 and p = 0.001, respectively). Alternatively, the charge/day and cost/day for each group were very similar. These calculations yielded a cost/day and charge/day of $2,859/day and $1,808 for the patients in group I (Figs 4, 5). These values were less than the $3,371/day and $2,071/day calculated for group II. However, these differences failed to reach significance (p = 0.24 and p = 0.16, respectively). The patients who were supported with a ventricular assist device (VAD) were evaluated further for the period from time of implantation until transplantation. This period yielded a charge/day and cost/day of $2,479 and $1,390 respectively, which was similar to the values for the pretransplantation interval.
|
|
|
|
). These results show that after transplantation 100% of the group I patients were alive at 1 month, and 81% were still alive at 6 and 36 months. The results from group II demonstrated that 92.5% of these patients were alive at 1 month; however, values at 6 and 36 months fell to 81.0% and 69.9%, respectively (Table 4). Log rank comparison of these survival curves demonstrated no significant differences between the two groups (p = 0.72).
|
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
Patients with prolonged waiting times are at risk for progressive deterioration of cardiac function with imminent secondary damage to multisystem end organs [1, 17, 18]. With failing cardiac function, these patients are likely to require admission for aggressive medical intervention such as inotrope therapy, ventilatory support, or temporary intraaortic balloon counterpulsation. Many times, these patients have irreversible cardiac decline and undergo chronic hospitalization until a suitable donor organ becomes available. Our local experience demonstrates the impact of a limited donor pool as recipients of a heart transplant at Pennsylvania State University since February 1993 have been exclusively status I at time of operation.
Continued decline of cardiac function despite maximal medical management has consequent end-organ deterioration, and these patients are at increased risk of transplant failure [12], or may die before a donor organ becoming available. The development of mechanical ventricular assistance has provided an additional means of supporting these patients, as well as reversing or preventing further end-organ damage. Early experience with mechanical assistance had limited success for varied reasons [6, 8, 19, 20]. Many of the earliest patients had delayed evaluations and underwent implantation of a VAD after suitable windows of success had passed. Additionally, different combinations of devices were used before consistent methods of artificial support, such as the exclusive use of Pierce-Donachy LVADs at our institution, were developed [15]. These experiences were mirrored through the published results of multiple institutions over this period [6, 8]. The implementation of stricter criteria and earlier evaluation of candidates for VAD insertion, as well as the benefit of nearly a decade of experience, has resulted in improved use of these devices as a bridge to transplantation. In addition, the increased use of these devices over time is emphasized by experience at our institution as shown in Figure 7. These plots demonstrate that one third of our patients who received transplants in 1992 and 1993 underwent LVAD assistance before transplantation, an increase from 7% in 1984 to 1985, despite an equal total number of transplantations for each period.
|
Despite the emphasis on costs of medical technology, differences between the clinical success rates of these treatment approaches provides additional insight for comparison. Although the patients receiving mechanical ventricular assistance have a unique set of risks including bleeding and neurologic sequelae [12], previous reports have alluded to the improved general health of these patients, physically and mentally, at time of transplantation [21–23]. The patients in group I had an improved transplantation rate, and 100% of the patients who underwent transplantation were discharged from the hospital. Although transplantation rates did not reach significance, the results clearly demonstrate the similarity between these modalities at supporting patients to transplantation. Further, the patients who underwent implantation of an LVAD had significantly improved rates of discharge from the hospital. Steps toward implanting these devices earlier in patients with deteriorating cardiac failure, as well as recent advances including the use of aprotinin and the development of completely implantable devices and improved blood interface surfaces, should serve to further improve outcome statistics of VAD-assisted patients [24, 25].
The scope of the financial implications of treating patients with cardiac failure may be appreciated by the size of this population of patients. However, there must be uniform comparison of the treatment modalities associated with end-stage cardiac disease. Our evaluation shows that the patients in group II had significantly reduced total charges/costs associated with the transplant admission. When these results were subjected to normalization for length of admission of each patient, cost/day and charge/day were determined for each patient. The mean values calculated for each group were quite similar. Although these results failed to reach significance, the patients supported with an LVAD before transplantation were found to have reduced daily expenses when compared with the medically treated population. Current management at the Pennsylvania State University entails chronic hospitalization in an intermediate care setting for the assisted patients, which is associated with a considerable daily bed charge. Recent experience at other institutions with outpatient care of these patients [22] suggests that these expenses can be reduced markedly and thus significantly affect the financial comparison between the assisted and medically treated patients. Additionally, the reduction in costs associated with improved transplantation and discharge rates and the impact of allowing these patients to return to some capacity of work while awaiting transplantation would only serve to increase the advantages of mechanical assistance.
Finally, complete comparison between treatment modalities must consider long-term follow-up results. All the patients undergoing heart transplantation currently have an expected 3-year survival of 74.3% [26]. Patients who underwent implantation of an LVAD had an increased 30-day survival, although survival approximated that of the medically treated group at 6 months. These results remained similar at 3 years as well. The survival results for both groups were similar to reported registry results [12, 26]. Previous studies have noted that after the period of operative mortality has passed, longer term survival becomes a reflection of problems inherent to cardiac transplantation rather than a function of pretransplantation treatment [12]. Despite improved survival rates at early intervals of follow-up for group I, these results failed to reach significance. These findings of equal long-term survival further demonstrate the success associated with mechanically assisted patients and emphasize the role of this treatment modality in the care of these critically ill patients.
In conclusion, over the last 10 years, there have been large advances in the implementation of VADs as a bridge to transplantation. Most notably patients are evaluated under stricter and better organized inclusion criteria before implantation of an assist device. This has had a twofold effect of limiting the institution of support in patients unlikely to benefit from ventricular assistance and encouraging consulting physicians to address the possibility of mechanical circulatory support earlier in a patient's course. Patients currently undergoing implantation of a VAD have significantly longer hospitalizations before undergoing transplantation when compared with the medically treated patients. Thus it is not surprising that these patients have increased total admission expenses. However, when costs are normalized to admission length, patients who are assisted mechanically and subsequently undergo transplantation have decreased daily costs/charges when compared with those treated medically. These differences will likely become significant as the mechanically assisted patients are discharged to outpatient care facilities. Additionally, continued research on completely implantable devices is likely to result in devices requiring less care, with decreased associated risks and morbidity and hence an expected additional reduction in cost of treatment. Further, discharging these patients increases the possibility of reintroducing these patients to the productive work force while awaiting transplantation. Therefore, these devices, with equal transplantation rates and increased discharge rates, may be associated with decreased expenses to maintain these patients before transplantation, particularly as this therapy is instituted earlier during the course of progressive heart failure. Finally, as the average duration of mechanical support increases secondary to a limited donor pool for patients with end-stage cardiac failure, the experience gained will have a concomitant significant impact on use of completely implantable devices. Although this study remains a single institutional experience, the addition of data from multiple centers to this early analysis should further strengthen these conclusions regarding mechanical support.
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
Related Article
This article has been cited by other articles:
|
|
 |
|
  F. Beyersdorf Economics of ventricular assist devices: European view Ann. Thorac. Surg., March 1, 2001; 71 (2007): S192 - S194. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. M. Portner Economics of devices Ann. Thorac. Surg., March 1, 2001; 71 (2007): S199 - S201. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. J. Bank, S. H. Mir, D. Q. Nguyen, R. M. Bolman III, S. J. Shumway, L. W. Miller, D. R. Kaiser, S. M. Ormaza, and S. J. Park Effects of left ventricular assist devices on outcomes in patients undergoing heart transplantation Ann. Thorac. Surg., May 1, 2000; 69(5): 1369 - 1374. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. L.S. Morales, K. A. Catanese, D. N. Helman, M. R. Williams, A. Weinberg, D. J. Goldstein, E. A. Rose, and M. C. Oz SIX-YEAR EXPERIENCE OF CARING FOR FORTY-FOUR PATIENTS WITH A LEFT VENTRICULAR ASSIST DEVICE AT HOME: SAFE, ECONOMICAL, NECESSARY J. Thorac. Cardiovasc. Surg., February 1, 2000; 119(2): 251 - 259. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. S. Couper, R. J. Dekkers, and D. H. Adams The logistics and cost-effectiveness of circulatory support: advantages of the ABIOMED BVS 5000 Ann. Thorac. Surg., August 1, 1999; 68(2): 646 - 649. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. C. Gelijns, A. F. Richards, D. L. Williams, M. C. Oz, J. Oliveira, and A. J. Moskowitz Evolving Costs of Long-Term Left Ventricular Assist Device Implantation Ann. Thorac. Surg., November 1, 1997; 64(5): 1312 - 1319. [Abstract] [Full Text]
|
|
|
|
 |
|
 M. G. Massad and P. M. McCarthy Will permanent LVADs be better than heart transplantation? Eur. J. Cardiothorac. Surg., April 1, 1997; 11(suppl): S11 - S17. [Abstract] [PDF]
|
|
|
|
|
|
W. E. Richenbacher and W. S. Pierce Mechanical Circulatory Support Ann. Thorac. Surg., November 1, 1996; 62(5): 1558 - 1559. [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 |
