| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1997;64:1312-1319
© 1997 The Society of Thoracic Surgeons
International Center for Health Outcomes and Innovation Research, Departments of Surgery and Medicine, and School of Public Health, Columbia University, College of Physicians and Surgeons, and The Presbyterian Hospital, New York, New York
Accepted for publication May 12, 1997.
 |
Abstract |
|---|
 TopFootnotes Abstract IntroductionBackgroundPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Methods. We used the ratio-of-cost-to-charges method to measure hospital costs and payments for physician time. We examined time trends in the resource use of 50 pneumatic left ventricular assist device recipients, using actuarial techniques and regression modeling.
Results. The average actual cost of left ventricular assist device support is $221,313 over an average of 9.5 months. If there had been no Food and Drug Administration regulatory policy precluding hospital discharge before 30 days, this value would have been $201,148. Based on this latter figure, the average predicted first-year cost is $219,139. The length of the intensive care unit stay, one of the most costly components of care, decreased significantly over time.
Conclusions. The high costs of left ventricular assist device implantation are similar to those reported for cardiac transplantation. Given their success in supporting survival, we anticipate that these devices will be similarly cost-effective. However, further research is imperative to determine the cost-effectiveness of these devices beyond the introductory phase, when costs, benefits, and Food and Drug Administration requirements have stabilized.
|
Introduction |
|---|
|
TopFootnotesAbstractIntroductionBackgroundPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Advances in medical therapy have had an important impact on the symptomatic status and short-term survival of patients with moderate to severe heart failure. However, existing pharmacologic agents have met with only moderate success in patients with class IV heart failure, and the 1-year survival rate is only 40% to 50% [7]. For these patients, cardiac replacement therapy in the form of cardiac transplantation is now the only viable treatment option. Cardiac transplant recipients have an in-hospital mortality rate of less than 5%, a 1-year survival rate approaching 85%, and a 5-year survival rate of around 65% [8]. Moreover, cardiac transplantation markedly improves quality of life, and it has been found to be a cost-effective alternative to medical therapy [9]. Yet, the success of cardiac transplantation remains limited by the complications of long-term immunosuppressive therapy, the development of allograft coronary artery disease, and, most importantly, the current serious shortage of donor organs. It has been estimated that at least 16,500 individuals per year in this country would be suitable candidates to receive donor hearts [10]. Despite major efforts to enhance donor heart procurement, however, the actual number of donor hearts harvested over the past few years has remained relatively constant at 2,000 annually [11]. Mortality for patients on the waiting list has risen, which has increased markedly interest in new treatment options for this disease [12].
Among the new therapeutic interventions that are currently under consideration are the left ventricular assist device (LVAD), cardiac myoplasty, and left ventricular reduction surgery. Of these, the LVAD perhaps has been used most widely. Since 1986, pneumatically driven LVADs have been used experimentally to sustain the lives of patients with end-stage heart failure who have become decompensated on maximal medical therapy while awaiting cardiac transplantation [13]. The success of this "bridging" experience led to the approval of the Thermocardio Systems, Inc (TCI, Woburn, MA) LVAD by the Food and Drug Administration (FDA) in September 1994. During this period, advances in LVAD technology led to the development of a more compact and portable, electric version of this device, which opened up the possibility of long-term, out-of-hospital use. Given the limitations of current treatment strategies for end-stage heart failure (both the lack of efficacy of pharmacologic agents and the unavailability of donor hearts), as well as the favorable experience with LVADs, there are increasing pressures to use these devices, not only as a bridge, but as an alternative to transplantation. In the current environment of constrained health care resources, it is imperative that this broadening of indications be guided by rigorous trials of the benefits and costs of this emerging technology.
In this study, we examined the costs—both inpatient and outpatient—associated with the electric LVAD at Columbia Presbyterian Medical Center (CPMC) over a 2-year period. We examined all rehospitalization costs and projected an average annual cost of care. Given that the LVAD is a new surgical intervention, the patterns of resource consumption can be expected to change significantly over time. To assess these dynamics, we reviewed our experience in 62 patients, starting with our first pneumatic implant in 1990. Thus, this study provides insight into some of the economic dimensions associated with the learning curve of a new surgical intervention.
|
Background |
|---|
|
TopFootnotesAbstractIntroductionBackgroundPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
A French study compared the costs of mechanical support in 6 patients (3 who received total artificial hearts, 1 who received an LVAD, and 2 who received biventricular devices) with those of pharmacologic support in 14 patients [15]. Using payments, the cost per patient at 1 month was $84,683 for those who received a mechanical assist device and $38,326 for those who received medical therapy alone. This experience included 5 transplantations (before 2 weeks) and 3 deaths in the 1-month costing. This study used payments-made as a surrogate for actual costs, which is known to be highly country-specific, that is, not easily translatable to other health care systems. Finally, the results are based on very few select patients, which means that the confidence interval for the cost estimates must be very broad and the generalizability of the results for use in other populations and venues may be limited.
The third study reviewed the pretransplantation cost experience of 43 patients, 12 of whom received the Pierce-Donachy (Penn State, Hershey, PA) LVAD and 31 of whom continued to receive pharmacologic support alone [16]. For a mean pretransplant hospital stay of 123.2 days, which included an average of 51.6 days of LVAD support, the average cost was $186,131 (charges totaled $302,048). By comparison, the pharmacologically managed patients remained in the hospital before transplantation for 52.6 days, at an average cost of $100,115 (charges totaled $165,219). A final study compared the hospital charges for 6 patients receiving conventional therapy with those of 6 patients receiving conventional therapy plus TCI LVAD support [17]. Partly as a result of FDA regulatory policies, the patients in the LVAD group had a significantly longer length of hospital stay (185 versus 51 days, respectively) and, therefore, higher total charges per patient ($435,133 versus $268,696, respectively). The authors of both articles acknowledge that the costs for the LVAD group would be considerably less if patients were allowed to go home instead of remaining in the hospital or in an intermediate care facility.
In summary, the costs of long-term LVAD support have not been addressed adequately in the clinical literature. Existing studies were not designed to capture long-term outpatient costs, including readmissions to the hospital for device- and disease-related events. Moreover, none of these studies address costs in a dynamic framework, taking into account the economic effects of improved expertise and modifications of FDA regulations. It is these issues that this paper intends to address.
|
Patients and Methods |
|---|
|
TopFootnotesAbstractIntroductionBackgroundPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
In this article, we review costing data obtained with TCI's pneumatically driven and electrically driven LVADs. The pneumatic HeartMate 1000 IP LVAD consists of a pusher-plate blood pump driven by a portable external console by a percutaneous driveline. In the VE, or "wearable," system, the pump remains the same, but the console has been replaced by a belt-mounted microcontroller and battery pack. A defining feature of both TCI devices is that sintered titanium microspheres and integrally textured polyurethane are used on their rigid metallic and flexing components. These textured surfaces promote the formation of a blood-compatible biologic lining, obviating the need for continuous heparin or warfarin anticoagulation. Despite the lack of systemic anticoagulation in most patients, few thromboembolic events (7%) were reported among TCI LVAD recipients [18].
Cost Data
Costing health care services is replete with practical and conceptual difficulties. Although analyses historically have used the charges that providers bill as proxies for the costs of resources, charges may differ substantially from the resource costs of delivering care. In contrast to charges, payments relate to actual financial transactions, but these fixed reimbursements do not specify the actual amount of resources used among the various hospital cost centers. The use of payments is hindered further by the lack of a Diagnosis-Related Group or Current Procedural Terminology code specific to LVADs. Resource costs, the actual expenses paid to obtain the resources used to deliver care, are the most desirable conceptually, but standard methodologies are still under development. The ratio-of-cost-to-charges method comes closest to approximating actual resource costs. In light of this, we used the ratio-of-cost-to-charges method as the primary approach for measuring hospital costs, and we used payments for other providers' expenses. We take a "health care budget" perspective in this analysis by calculating the costs of all services associated with providing health care to LVAD recipients, regardless of who bears the cost.
We audited the hospital patient management system and gathered the line-item bills for hospital services and supplies for all 12 LVAD recipients being analyzed. For each patient, we captured the billable item, the date of the charge, and the amount of the charge. Inpatient routine charges were broken down into regular floor days and special care days (surgical cardiac intensive care unit [ICU], coronary care unit, and stepdown unit). For ancillary services billed, each item was categorized into a departmental category (eg, chemistry, radiology) and each department category then was categorized further into a utilization category (eg, diagnostic tests) (Table 1
). For each patient, we summed the total charges in each departmental category that were incurred during the period ranging from the day of implantation to the date of hospital discharge. For each departmental category, we multiplied the total charges by the corresponding ratio of cost to charges submitted by CPMC in its Health Care Financing Administration 1994 and 1995 Institutional Cost Reports.
|
Outpatient services included physician care, diagnostic tests, and medications. The costs of physicians' services were approximated by the reimbursement received. All diagnostic tests were performed in the hospital and the associated charges again were converted to costs using hospital ratios of cost to charges. Outpatient medications were costed using Medicare's listed price for each drug.
All readmissions were included in the analysis except one. This readmission concerned a patient who needed a replacement controller component, which is performed routinely in the outpatient setting. Unfortunately, such a controller was unavailable, and the patient had to receive a pneumatic device, which, as a result of FDA regulation, necessitated continued hospitalization until transplantation. Readmissions were costed in the same manner as implantation admissions.
Discharge Criteria
At the time of study, FDA regulations required that all patients who received the TCI VE LVAD remain in the hospital for at least 30 days after implantation before being discharged home. For each patient, we determined when that patient could have been released from the hospital if there were no regulatory restrictions. We measured the following factors to arrive at this judgment:
Physical performance tests were administered to assess the patients' independence with simulated activities of daily living. To qualify for hospital discharge, patients had to be capable of writing, eating, lifting, dressing, bending, walking, and climbing stairs. We calculated costs based on the actual length of the hospital stay, as well as based on the shorter, "clinically sufficient" hospital stay.
Statistical Analyses
We report the mean cost and standard deviation for each resource utilization category, both inpatient and outpatient. Average annual hospitalizations and costs were calculated by determining the average number of hospitalizations and associated costs per patient-day of LVAD support and projecting this to 1 year. Average weekly outpatient costs during the observation period were projected to 52 weeks. To examine time trends in resource utilization over a 5-year period (1990 to 1995), we used Kaplan-Meier actuarial techniques (intensive care unit stays) and linear regression modeling. Statistical significance was determined using the log-rank test of the heterogeneity between groups (as defined in the Results section) and t tests for the significance of regression parameter estimates.
|
Results |
|---|
|
TopFootnotesAbstractIntroductionBackgroundPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
provides a breakdown of patient characteristics). The outcomes for this population during the period of study included 2 deaths, 8 transplantations, and 2 transplant candidates with device support. The average number of LVAD-supported days was 177 days, with a range of 13 (due to perioperative mortality) to 481 days (remaining on LVAD support). The average length of hospital stay was 43.5 days.
|
. As the table indicates, we have distinguished actual costs from those that are clinically sufficient. Given FDA regulatory policies, VE LVAD recipients have been delayed in leaving the hospital. On the basis of objective clinical criteria, such as absence of infections and satisfactory ambulation (see Methods section), we designated when patients could have gone home if there had been no regulatory restrictions. The projected average (± standard deviation) clinically sufficient length of stay for the 12 VE LVAD recipients included in this analysis would have been 17.5 ± 5.32 days. The mean cost of care over the "clinically sufficient" period was $74,202 not including cost of the device and $141,287 including the cost of the device.
|
compares the resource consumption categories for the actual versus the clinically sufficient hospital stay as a percentage of total costs. Note that the average cost per day of the actual hospitalization is much lower than that of the clinically sufficient course ($4,225 and $8,058, respectively, including the device). This reflects the lack of clinical activities beyond the clinically sufficient hospital stay period. Generally, the only charges listed for patients beyond this period were bed charges, diagnostic and laboratory tests, and rehabilitation. Given that these diagnostic and laboratory tests would have been conducted during the weekly outpatient visits if the patients had been discharged, the more than $16,000 difference in bed charges accrued between the clinically sufficient and actual hospitalizations is by far the greatest.
|
). The average professional payment per visit was $128. There were a total of 11 readmissions during the period of observation, which involved 5 patients and 127 hospitals days. The total cost of readmissions was $282,178, which corresponds to an average of $56,436 per readmitted patient and $25,653 per readmission. The breakdown of the readmission costs per resource category and the reasons for readmission are depicted in Table 5.
|
|
Projecting Annual Costs
To capture better the cost of long-term LVAD use in patients who would not qualify for transplantation, we made the following projection for the first year, based on the experience of the 6 patients who were discharged, after implantation, during 1995. This projection includes the costs of the initial hospitalization (clinically sufficient period), readmissions, and outpatient treatment. Of a total of 2,883 LVAD-supported days, patients were readmitted for 127 days, which amounted to 8.5% of the total outpatient period. Using these percentages, the projected annual cost of LVAD support would be $219,139.
Trends Over Time: Pneumatic and VE LVADs
Given that this is an emerging procedure, the costs can be expected to fall over time. To explore time trends, we examined selected categories of resource consumption for all LVAD recipients over a 5-year period. This examination involved the 12 VE LVAD recipients mentioned earlier, as well as 50 pneumatic LVAD recipients, starting with the first patient who underwent implantation in 1990 (Table 2
). Right ventricular failure is an important variable that is highly predictive of overall mortality and cost. We examined the outcome of right ventricular failure, which resulted in death, the need for a right ventricular assist device, or the need for 2 weeks of treatment with dobutamine. There were 4 cases (40%) of right ventricular failure among the first 10 patients, compared with 1 case (10%) among the last 10 patients. Although it is clinically important and likely resulted from the use of nitrous oxide in the perioperative care of our patients, this trend is not statistically significant.
We used regression models to correlate the ICU length of stay with the experience (as defined by time since inception) of our LVAD program. Adjusting for patient age and etiology of heart disease, program experience was correlated inversely with ICU stay (ßstudy-time coefficient = -0.026, p < 0.016; r2 = 0.964). Given that the ICU stay is among the most costly components of the implantation hospital stay, we anticipate a cost reduction related to further growth in institutional skill and experience.
|
Comment |
|---|
|
TopFootnotesAbstractIntroductionBackgroundPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
We reviewed the resource implications of implanting these devices, including the costs of the initial hospitalization, readmissions, and outpatient care. The projected annual cost of LVAD implantation is $219,139, of which the initial hospitalization accounts for about 64% ($141,287). Within the initial hospitalization period, the device is by far the single largest cost component. Excluding professional payments, the time spent in the intensive care unit is the next largest contributor to resource consumption, followed by the diagnostic tests performed.
Because the LVAD is an emerging technology, we can expect that outcomes, costs, and indications for use will change dramatically over time. With technologic change, a reduction in manufacturing costs, and increasing competition from other manufacturers, it is likely that the price of the device will decrease sharply. A similar dynamic can be expected regarding the length of the hospital stay. Even during our relatively short experience with this device, we were able to show a significant trend toward reducing the postoperative stay in the intensive care unit. Soon, we anticipate that FDA regulations concerning the hospital stay after implantation will become less stringent and further reduce the implantation costs. Similarly, as the technology becomes more established, less demanding regulatory requirements for initial and follow-up laboratory testing can be expected, further contributing to a reduction in overall costs.
A major question is whether the use of this device as a bridge to transplantation will constitute a cost-effective treatment. The answer to this question depends on the effectiveness as well as the costs of the LVAD as compared with cardiac transplantation alone. Early data from the TCI bridge trial, a comparative but nonrandomized study, suggest a substantial improvement in survival (more than 70% of patients who were bridged survived to transplantation, compared with 36% of those who were not bridged). Moreover, our studies have demonstrated that these patients are capable of resuming their lives with relatively few limitations. Twenty-five percent of patients have returned to work or school while awaiting transplantation, and objective measures of functional capacity, quality of life, and patient preferences have been exceptionally high in comparison with other chronic diseases [19, 20]. On the cost side, the introduction of an additional procedure obviously adds significantly to the overall expenditures for transplantation. Whether this increased cost balances the survival gain over transplantation alone requires a rigorous cost-effectiveness analysis, which we hope to address soon.
In conducting such an analysis, one must take into consideration that the outcome and costs parameters of both LVAD implantation and transplantation will change. For example, with the growing number of transplant candidates and the limited number of donor hearts, the waiting list for heart transplantation continues to increase substantially. In 1988, the median waiting time for hearts was 116 days; by 1994, the waiting period had increased, by more than 30%, to 184 days [21]. This means that the length of the hospital stay before transplantation is increasing, which dramatically increases the cost of supporting transplant candidates. At some point, the substantial initial costs of implanting an LVAD may be counterbalanced by the additional costs of hospitalizing transplant candidates awaiting their hearts. Should that point be reached, it is conceivable that bridging with the LVAD will be not only cost-effective, but cost-saving. Of course, in the long run, the emergence of xenotransplantation and new methods of immunosuppression might cause the cost-effectiveness ratio to go the other way.
Ultimately, the LVAD will be considered, not as a bridge, but as a stand-alone technology. In fact, we currently are conducting a randomized trial, including a cost-effectiveness analysis, comparing LVADs with medical management in patients with end-stage heart failure who are not eligible for transplantation. In this trial, we will examine both the initial and the long-term costs and benefits of both medical care and LVAD therapy. On the device side, particular emphasis will be given to the costs of device complications, including the need for device replacement because of failure. As a result of our current experience, we know that the cost of implanting an LVAD is very similar to the cost of transplanting a heart ($117,352 versus $102,828, respectively, without professional fees) for the initial hospitalization [22]. If the assumptions of this trial bear out in terms of increased survival (ie, a 33% to 50% reduction in the mortality rate) and improved quality of life, one can expect a favorable cost-effectiveness ratio. This would expand the indications for use, including the possibility of using the LVAD as an alternative to transplantation. The resulting increase in the use of this cost-effective intervention will lead to substantial aggregate expenditures for society.
In fact, this experience is far from unique, as has been illustrated recently with laparoscopic cholecystectomy, coronary artery bypass grafting, and percutaneous transluminal coronary angioplasty. The introduction of laparoscopic cholecystectomy, for instance, significantly reduced the pain associated with open repairs as well as the per-patient cost, by 25% [23]. Nevertheless, total expenditures for a large health maintenance organization rose by 18%. The reason was simple: associated with the 25% cost reduction per patient was an increase in the number of gallbladder operations by no less than 60% [23]. We suggest that this phenomenon may foreshadow the future economics of medical care in affluent societies, reinforced by the aging of their populations: expectations of new technologies offering the prospect of expenditure reduction are likely to continue to be frustrated for the excellent reason that the quality of medical care also is likely to continue to improve.
|
Acknowledgments |
|---|
|
TopFootnotesAbstractIntroductionBackgroundPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
Footnotes |
|---|
|
TopFootnotesAbstractIntroductionBackgroundPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopFootnotesAbstractIntroductionBackgroundPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. A. Murray, S. Osaki, N. M. Edwards, M. R. Johnson, J. L. Bobadilla, E. A. Gordon, M. Sanderfoot, and T. Kohmoto Multidisciplinary approach decreases length of stay and reduces cost for ventricular assist device therapy Interactive CardioVascular and Thoracic Surgery, January 1, 2009; 8(1): 84 - 88. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Beyersdorf Economics of ventricular assist devices: European view Ann. Thorac. Surg., March 1, 2001; 71 (2007): S192 - S194. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. J. Moskowitz, E. A. Rose, and A. C. Gelijns The cost of long-term LVAD implantation Ann. Thorac. Surg., March 1, 2001; 71 (2007): S195 - S198. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. M. Portner Economics of devices Ann. Thorac. Surg., March 1, 2001; 71 (2007): S199 - S201. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. C. Deng, M. Loebe, A. El-Banayosy, E. Gronda, P. G. M. Jansen, M. Vigano, G. M. Wieselthaler, B. Reichart, E. Vitali, A. Pavie, et al. Mechanical Circulatory Support for Advanced Heart Failure : Effect of Patient Selection on Outcome Circulation, January 16, 2001; 103(2): 231 - 237. [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]
|
|
|
|
|
|
E. A. Rose, A. J. Moskowitz, M. Packer, J. A. Sollano, D. L. Williams, A. R. Tierney, D. F. Heitjan, P. Meier, D. D. Ascheim, R. G. Levitan, et al. The REMATCH trial: rationale, design, and end points Ann. Thorac. Surg., March 1, 1999; 67(3): 723 - 730. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Westaby, T. Katsumata, R. Houel, R. Evans, D. Pigott, O. H. Frazier, and R. Jarvik Jarvik 2000 Heart : Potential for Bridge to Myocyte Recovery Circulation, October 13, 1998; 98(15): 1568 - 1574. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
