Biology of Blood and Marrow Transplantation
Volume 14, Issue 2 , Pages 197-207, February 2008

Costs of Allogeneic Hematopoietic Cell Transplantation with High-Dose Regimens

  • Akiko M. Saito

      Affiliations

    • Division of Medical Oncology, Dana-Farber Cancer Institute, and Department of Medicine, Brigham and Women's Hospital, Boston, Massachusetts
    • Corresponding Author InformationCorrespondence and reprint requests: Akiko M. Saito, MD, PhD, Department of Medical Oncology, Dana-Farber Cancer Institute, 44 Binney Street, SM 277, Boston, MA 02115, USA, Phone: 617-632-6637, Fax: 617-632-3161.
    • ,
  • Corey Cutler

      Affiliations

    • Division of Medical Oncology, Dana-Farber Cancer Institute, and Department of Medicine, Brigham and Women's Hospital, Boston, Massachusetts
    • ,
  • David Zahrieh

      Affiliations

    • Department of Biostatistics and Computational Biology, Dana-Farber Cancer Institute, Boston, Massachusetts
    • ,
  • Robert J. Soiffer

      Affiliations

    • Division of Medical Oncology, Dana-Farber Cancer Institute, and Department of Medicine, Brigham and Women's Hospital, Boston, Massachusetts
    • ,
  • Vincent T. Ho

      Affiliations

    • Division of Medical Oncology, Dana-Farber Cancer Institute, and Department of Medicine, Brigham and Women's Hospital, Boston, Massachusetts
    • ,
  • Edwin P. Alyea

      Affiliations

    • Division of Medical Oncology, Dana-Farber Cancer Institute, and Department of Medicine, Brigham and Women's Hospital, Boston, Massachusetts
    • ,
  • John Koreth

      Affiliations

    • Division of Medical Oncology, Dana-Farber Cancer Institute, and Department of Medicine, Brigham and Women's Hospital, Boston, Massachusetts
    • ,
  • Joseph H. Antin

      Affiliations

    • Division of Medical Oncology, Dana-Farber Cancer Institute, and Department of Medicine, Brigham and Women's Hospital, Boston, Massachusetts
    • ,
  • Stephanie J. Lee

      Affiliations

    • Division of Clinical Research, Fred Hutchinson Cancer Research Center, Seattle, Washington

Received 11 September 2007; accepted 25 October 2007.

Article Outline

Abstract 

To characterize the costs of allogeneic hematopoietic cell transplantation with high-dose regimens (HDCT), we analyzed clinical information and costs of 315 HDCT recipients during a 4-year study period beginning in 2000. Multivariate analyses were performed to identify pre- and/or post-HDCT factors predicting higher costs within the first year. Overall survival (OS) at 100 days and 1 year were 80% and 58%, respectively. The median cost and days of hospitalization were $102,574 in 2004 US dollars and 36 days in the hospital for 100 days, and $128,800 and 39 days in the hospital for 1 year. Early costs, defined as costs within the first 100 days, accounted for 84% of total costs within the first year. Inpatient costs comprise 94% of the early costs, but only 61% of the later costs defined as costs incurred between 101 days and 1 year. Of the pre-HDCT factors, unrelated donors and advanced disease risk were significantly associated with increased cost. When post-HDCT events were also considered, these pre-HDCT factors were no longer independently predictive of high cost. Instead, severe complications post-HDCT were associated with higher costs, increasing total costs $20,228 on average. If no complications occurred, the mean cost within the first year was $79,222. These results provide cost estimates for complicated and uncomplicated HDCT procedures, as well as costs for management of specific transplant complications.

Key Words: Costs, Outcomes research, Hematopoietic cell transplantation with high-dose regimens, Hematological malignancies

 

Back to Article Outline

Introduction 

Since the first successful allogeneic hematopoietic cell transplantation (HCT) was performed almost 40 years ago [1], HCT with high-dose regimens (HDCT, previously called “myeloablative transplantation”) has become an established therapy for patients with hematological diseases. However, relapse or serious complications related to HDCT, including graft-versus-host disease (GVHD), sepsis, cytomegalovirus (CMV) infection, and fungal infection still limit the success of the procedure 2, 3, 4, 5, 6, and also increase the financial cost of HDCT. Indeed, HDCT is recognized as a very costly procedure, ranging from approximately $30,000 for an uncomplicated autologous HDCT to $200,000 for an allogeneic HDCT using an unrelated donor [7]. A 2003 National Health Interview Survey data showed that 17% of Americans younger than 65 years lack health insurance, and 24% of those 65 years and older have only Medicare coverage [8]. Few people can afford HDCT as an out-of-pocket expense. Even for insured persons, high costs related to HDCT may limit access to transplantation because of coverage denial, high deductibles, or lifetime payout caps 9, 10. Therefore, the identification of factors associated with the high cost of HDCT and the identification of methods for reducing those costs without compromising clinical outcomes are of considerable significance.

There is a paucity of literature looking at costs associated with pre-HDCT characteristics or specific complications post-HDCT 11, 12, 13, 14, and all of these studies evaluated patients who received HDCT before 2000. Previous studies identified a variety of possible cost drivers: level of experience of each transplant center 15, 16, 17; the health care system[18]; year of transplantation [14]; conditioning regimen or GVHD prophylaxis 19, 20; changes in diagnostic tools or supportive care 21, 22, 23; clinical outcomes [17]; disease; disease status; donor characteristics; the graft source; the age and the health status of recipients; economic status of each country; governmental economic policies [14], and others. A number of improvements in HDCT technology have occurred since 2000. In addition, patients who previously would have undergone risky HDCT because of disease indications may instead be receiving HCT with reduced intensity regimens (RIC). There is only one European study evaluating the treatment cost of severe acute GVHD (aGVHD) for patients including those transplanted after 2000 [24]. However, half of the patients in that study received RIC; characteristics of toxicities and costs after RICT were considerably different from those of HDCT, despite similar clinical outcomes [25]. Thus, we evaluated the cost of HDCT in recent years. We compared our results with those from published literature to determine if recent practice changes have decreased the cost of HDCT or, at least, changed the spectrum of cost drivers.

Back to Article Outline

Materials and Methods 

Study Population 

Between June 2000 and July 2004, a total of 376 patients with hematological malignancies received a HDCT at the Dana-Farber Cancer Institute / Brigham and Women's Hospital (DFCI/BWH). The analysis excluded some patients: those who did not receive high-dose cyclophosphamide and fractionated total body irradiation (n=12); those who received cord blood grafts (n=1) or both bone marrow (BM) and peripheral blood grafts (n=1); and those who had undergone allografting previously or, following the index, HDCT within a year (n=10). Patients were also excluded if nontransplant costs were included in the hospitalization (n=11 with prolonged pre-HDCT hospital stays for reasons not directly related to their transplantation) or if cost data were incomplete (n=27). Pre-HDCT characteristics and clinical outcomes of these excluded patients were similar to those included in this study (data not shown). A total of 315 patients were included in the analysis.

The Institutional Review Board at the DFCI/BWH approved these cost studies, and all patients provided signed, informed consent for their HDCT procedures.

Conditioning Regimen, GVHD Prophylaxis, and Supportive Care 

Conditioning before HDCT consisted of high-dose cyclophosphamide and fractionated total body irradiation (TBI) with or without antithymocyte globulin (ATG). GVHD prophylaxis included cyclosporine or tacrolimus, with or without standard methotrexate (MTX) (15mg/m2 on day 1, 10mg/m2 on days 3, 6, and 11); tacrolimus, sirolimus, plus low dose MTX (5mg/m2, days 1, 3, 6, and 11); tacrolimus, plus sirolimus, without MTX; or T cell depletion (TCD). Stem cell source was either bone marrow or granulocyte colony stimulating factor (G-CSF) stimulated peripheral blood stem cells (PBSC). The assignment of conditioning regimen, GVHD prophylaxis, and/or graft source was based on protocols or clinical decisions.

The day of graft infusion was designated day 0. Patients received standard prophylactic acyclovir from day 5 until 1 year and Pneumocystis jiroveci prophylaxis with atovaquone, dapsone, or trimethoprim-sulfamethoxazole. Blood was obtained weekly after engraftment for CMV testing, and patients were treated pre-emptively with ganciclovir or valgancyclovir if clinically indicated. Levofloxacin was used as bacterial prophylaxis if specified by protocol. Patients were treated with broad spectrum antibiotics at the time of their first neutropenic fever, and with antifungal agents if applicable.

Patients were admitted to receive their conditioning chemotherapy and remained hospitalized until neutrophil engraftment, adequate oral intake, and an absence of uncontrolled medical problems. No patients underwent outpatient transplantation. Patients were usually provided post-HDCT care at the DFCI/BWH as described elsewhere [25]. If re-admission was necessary, they were usually re-admitted to the DFCI/BWH or transferred to DFCI/BWH shortly after re-admission at their local hospitals. Patients continued to be seen at DFCI/BWH through 1 year post-HDCT or until their deaths, even when also seen outside DFCI/BWH by their referring physicians. At the DFCI, the inpatient unit at the BWH can deliver intensive care unit (ICU) level care if needed, including staffing and equipment. In contrast to other programs, patients are not “transferred” to the ICU.

Clinical Data and Definition 

Data regarding baseline patient and transplant characteristics, complications, and relapse occurring within a year after HDCT were obtained from the clinical transplant database or patients' records. The “early” post-transplant phase was defined as being from initial admission to 100 days after HDCT; the “later” phase was defined as being from 101 days to 1 year after HDCT. The early and later phases were examined separately, using any complications or relapses occurring within each phase as potential predictors.

1) Disease status 

Disease status prior to HDCT was categorized into two groups. Standard risk was defined as: acute leukemia in first remission; chronic myelogenous leukemia (CML) in chronic phase; lymphoma in first remission; or refractory anemia without excess blasts. All other stages and types of hematological cancers were considered advanced risk.

2) Complications after HDCT 

Neutrophil recovery was defined as a neutrophil count of at least 0.5 x 109/L. Early neutrophil engraftment was defined as engraftment within 15 days after graft infusion. Platelet recovery was defined as a platelet count of at least 20 x 109/L. The diagnosis of aGVHD was based on clinical findings and/or biopsy of the skin, digestive tract, or liver and graded according to consensus definition[26]. Chronic GVHD (cGVHD) was also graded according to previously published criteria[27]. Veno-occlusive disease (VOD) [28], idiopathic pneumonia (IP)[29], and diffuse alveolar hemorrhage (DAH) 30, 31 were diagnosed by the attending physician using clinical criteria. Renal/bladder, neurological, or cardiac toxicity were graded according to the National Cancer Institute common toxicity criteria with grade 3 or more defined as severe organ toxicity. Severe infections were captured if patients received antibiotic, antiviral, and/or antifungal agents intravenously regardless of whether or not an organism was confirmed. In-hospital death was included as a potential predictor of costs, recognizing that death might be caused by one of the identified complications. However, the analysis was repeated excluding in-hospital death as a potential predictor, and results were similar.

3) Relapse 

Relapse was diagnosed based on hematologic parameters, tissue biopsy, bone marrow biopsy findings, and cytogenetic or molecular methods.

Costs and Hospital Stay 

Inpatient costs from admission for HDCT to 1 year post-transplant and outpatient costs during the same period were obtained from the DFCI/BWH's accounting system. Inpatient costs include: room costs; pharmacy; blood bank including stem cell infusion; laboratory tests; radiation therapy; and miscellaneous costs. Costs of donor identification and graft procurement were excluded because unrelated donor stem cell procurement is much more costly than obtaining cells from family members, and also because we did not have access to family member graft procurement costs [32]. In addition, non-medical costs such as caregiver time, transportation, and local housing costs were not included because they could not be collected retrospectively. We did not capture costs outside of the DFCI/BWH's accounting system. Our previous study documented that external clinic visits and hospital days outside of the DFCI/BWH system account for fewer than 4% of total costs within the first year post-HDCT [25]. Review of 50% of patients randomly selected from the present study cohort confirmed the low external costs (outside hospitalization costs/total costs within a year=3.1%). Costs were estimated using a relative value units methodology [33] within the hospital accounting system, in which costs were calculated by applying unit costs assigned to each health care service item to each patient's resource utilization. All costs were adjusted for inflation to the year 2004 using the medical care component of the consumer price index [34]. This cost analysis is from the perspective of the health care system rather than a societal perspective [35].

Initial hospital stay is defined as the period from the date of initial admission for conditioning, followed by graft infusion, to the first date of discharge. Subsequent hospitalizations within a year after transplantation were defined as subsequent hospital stays.

Statistical Analysis 

1) Clinical outcomes 

Time to post-transplant events was calculated using the Kaplan-Meier method, and comparisons were made by using the log-rank tests [36]. Multivariable logistic regression was used to investigate risk factors that were associated with development of post-transplant events. All analyses were performed on data collected by January 2006.

2) Evaluation of costs after HDCT 

We evaluated the total cost within the first year and the characteristics of cost breakdowns by phase after HDCT. To identify the pre-HDCT factors influencing high costs during the first year, a multiple linear regression model was created. The model included all the baseline patient and transplant characteristics: GVHD prophylaxis; patient age (>50 years versus [vs] ≤50 years); donor sex; donor type; patient and donor cytomegalovirus serological status; disease (acute lymphoid leukemia vs myeloid malignancy vs non-Hodgkin lymphoma [NHL]); disease status (advanced risk vs standard risk); graft source; and year of transplantation. To evaluate the impact of clinical events on cost, the presence of specific post-HDCT complications were included in the model. These complications were: slow recovery of neutrophils (neutrophil engraftment after 15 days or no recovery); morbidity such as VOD, IP/DAH, infection, more than or equal grade III organ toxicity, such as renal/bladder toxicity, neurological toxicity and cardiac toxicity, grade II to IV aGVHD, extensive cGVHD, and in-hospital death; and relapse during the first year. Separate analyses of costs were performed by the phase (ie early or later) after HDCT because only patients who survived more than 100 days require later costs. In addition, some complications occurring in the early phase, such as slow engraftment and VOD, are unlikely to contribute to later costs. Similarly, cGVHD occurring in the later phase should not contribute to early costs. Also, within the first year after transplant, all events were ascertained and all surviving patients included in the early and later phase models. Thus, our analytical methodology takes into account the timing of complications when assessing their relationship to cost.

Because this analytic approach is based on the assumption of a normal distribution of the data, we used the natural logarithm of the costs and excluded six patients with extreme outlier costs (>$450,000) within a year after HDCT to stabilize the variance in the multivariate analysis. We confirmed the same results by including these 6 patients. Results are transformed back to their original scale for presentation. We presented “incremental costs (IC)” and “ratios” in the results section. The IC is the estimated cost difference in dollars between patients who have a specific baseline characteristic, experience a specific complication, or relapse and those who do not. Costs of the two types of patient groups can be written in the form of a ratio, which is called the “ratio.” Because of the high, positive correlation between the days of hospital stay and costs (r = .88, P < .01), we excluded the days of hospital stay from the baseline and the full models in the multivariate analysis.

We investigated all the models that included 2 of the 9 possible baseline factors and the two-way interactions. Significant interaction terms (P < .05) were then included both in the baseline and the full models. All statistical tests were two-sided. P-values of less than .01 were considered significant.

Back to Article Outline

Results 

Patient and Transplant Characteristics 

Patient and transplant characteristics are summarized in Table 1. The median age of patients was 42 years; for donors, the median age was 38 years. Patients over 50 years are more likely to receive HDCT from older donors (38% vs 8%, P < .01). Although MTX in combination with a calcineurin inhibitor is our standard GVHD prophylaxis regimen, since 2000, 3 sequential phase II studies have evaluated a sirolimus-based regimen 37, 38, 39. In addition, patients who underwent transplantation earlier were more likely to have bone marrow (BM) grafts (P < .01). As a result, patients receiving MTX in their GVHD prophylactic regimen were more likely to undergo transplantation in earlier years (P < .01) and to have BM grafts than those who did not receive MTX (BM graft; standard dose MTX vs low-dose MTX vs no MTX: 64% vs 73% vs 18%, P < .01). Patients who received low-dose MTX as a GVHD prophylaxis on a research protocol were more likely to have HLA mismatched donors (standard dose MTX vs low-dose MTX vs no MTX: 14% vs 30% vs 2%, P < .01) and underwent transplantation before 2003. Median follow-up of all patients was 426 (interquartile range (IQR), 127-1062) days. Median follow-up of surviving patients was 1096 (IQR 741-1433) days.

Table 1. Patient and transplantation characteristics
VariablesN (%) [Interquartile range]
Number of patients315
Patients' age, median years42 [33-48]
Donors' age, median years38 [29-45]
Donor sex
Male184 (58)
Female131 (42)
HLA matching
Matched273 (87)
Mismatched42 (13)
Patient-donor relationship
Related donor158 (50)
Unrelated donor157 (50)
CMV serological status
CMV positive donor and/or patient170 (54)
CMV negative donor and patient145 (46)
Disease
ALL44 (14)
AML107 (34)
MDS45 (14)
CML64 (20)
NHL37 (12)
Other∗18 (6)
Disease status†
standard risk108 (34)
advanced risk207 (66)
Stem cell source
bone marrow152 (48)
peripheral blood163 (52)
GVHD prophylaxis
Cyclosporine or Tacrolimus with MTX107 (34)
Cyclosporine or Tacrolimus without MTX5 (2)
Tacrolimus with Sirolimus60 (19)
Tacrolimus, Sirolimus, and MTX82 (26)
TCD61 (19)
Year of transplant
200059 (19)
200187 (28)
200279 (25)
200372 (23)
200418 (6)
Days of median follow-up426 [127-1062]

ALL indicates acute lymphoblastic leukemia; AML, acute myelogenous leukemia; MDS, myelodysplastic syndrome; CML, chronic myelogenous leukemia; NHL, non-Hodgkin lymphoma; GVHD, graft-versus-host disease; MTX, methotrexate; TCD, T cell depleted transplantation

∗Others include chronic lymphoblastic leukemia, Hodgkin lymphoma, multiple myeloma, and other leukemia

†Standard disease risk was defined as: acute leukemia in first remission, chronic myelogenous leukemia in chronic phase, lymphoma in first remission, or refractory anemia without excess blasts. All other stages and types of hematological cancers were considered advanced disease risk.

Clinical Outcomes 

Table 2 summarizes aggregate clinical outcomes within the first year after HDCT. In the early phase, unrelated donor recipients experienced more complications including VOD (15% vs 8%), IP/DAH (15% vs 6%), severe infection (44% vs 28%), severe renal/bladder toxicity (19% vs 6%), severe neurological toxicity (15% vs 7%), and in-hospital death (22% vs 11%). Incidence of grade II to IV aGVHD (unrelated vs related: 39% vs 32%, P=.18) and relapse rate (unrelated vs related: 7% vs 11%, P = .60) did not differ between related and unrelated donor transplantation in the early phase. In the later phase, incidence of cGVHD for unrelated donor recipients was higher (70% vs 57%, P=.04), but that of extensive cGVHD (unrelated vs related: 53% vs 42%, P=.09), incidence of other post-HDCT complications and rate of relapse were similar between the two groups. Consequently, unrelated donor recipients had lower disease-free survival (DFS) and overall survival (OS) at 100 days and 1 year (DFS: 68% vs 82% at 100 days, 48% vs 59% at 1 year, P = .02; OS: 73% vs 87% at 100 days, 52% vs 64% at 1 year, P < .01). Several clinical factors were found to be associated with the development of complications. Minimization of MTX in the GVHD prophylactic regimen (no MTX, odds ratio (OR)=0.47, P < .01; low MTX, OR=.06, P = .03) and use of peripheral blood as the stem cell source (OR=.17, P < .01) were associated with early recovery of neutrophils. Patient age greater than 50 years was associated with later neutrophil recovery (OR=2.70, P = .01) and higher occurrence of IP/DAH (OR=2.30, P = .04). A positive CMV serology in donor or patient was associated with an increased risk of VOD (OR=2.77, P = .01). Grade II to IV aGVHD was associated with less relapse, both throughout the entire follow-up period (OR=0.58, P = .05) and when the analysis was limited to patients surviving more than a year (OR=0.39, P = .02). Prior grade II to IV aGVHD was also associated with extensive cGVHD in the later phase (OR=2.37, P < .01). Advanced risk disease was associated with a higher relapse rate both in the early phase (OR=2.32, P = .08) and in the later phase (OR=2.51, P = .04).

Table 2. Clinical outcomes
VariablesEarly phase N (%)Later phase N (%)
Number of evaluated patients315252
Post-transplant events
Neutrophil engraftment (≥0.5x109/L), median days [interquartile range]15 [13-18]-
Platelet engraftment (≥2.0x109/L), median days [interquartile range]25 [16-44]-
Veno-occlusive disease35 (11)-
Idiopathic pneumonia / diffuse alveolar hemorrhage33 (11)10 (4)
Infection (≥grade 3)110 (35)66 (26)
Viral38 (12)29 (12)
Bacterial79 (25)35 (14)
Fungal26 (8)13 (5)
Renal/bladder toxicity (≥grade 3)48 (15)23 (9)
Cardiac toxicity (≥grade 3)22 (7)19 (8)
Neurological toxicity (≥grade 3)35 (11)22 (9)
In-hospital death53 (17)45 (18)
Grade II-IV aGVHD in the early phase112 (35)-
Extensive cGVHD in the later phase-119 (47)
Relapse25 (8)37 (15)
Disease-free survival at 100 days or at 1 year75±2 %53±3 %
Overall survival at 100 days or at 1 year80±2 %58±3 %

Early phase indicates initial admission through 100 days after transplantation; later phase, from 101 days to 1 year (365 days) after transplantation; GVHD, graft-versus-host disease; aGVHD, acute GVHD; cGVHD, chronic GVHD.

Costs and Days of Hospital Stay 

Univariate results for costs and days of hospital stay are shown in Table 3. The median total costs within 100 days and 1 year after HDCT were $102,574 and $128,800, respectively. Initial hospitalization costs accounted for 79% of total costs in the early phase post-HDCT, and costs of the early phase accounted for 84% of all costs within the first year. Breakdown analyses of cost categories by post-transplant phase are shown in Figure 1A. In the early phase, inpatient costs were 94% of total costs; in the later phase, they accounted for 61% of total costs. Room costs were the largest category of costs, followed by pharmacy and blood bank, accounting for 76% of total costs in the early phase. Outpatient costs accounted for 6%, on average, in the early post-transplant period, increasing from 2.0% in 2000 to 7.1% in 2004. Similar patterns of cost distribution were seen regardless of whether patients developed complications or the types of complications.

Table 3. Univariate analyses of costs and days of hospital stay
VariablesMedianInterquartile range
Total cost, 2004 $
Total cost within the first 100 days102,57478,679-161,990
Total cost of the first year128,80090,511-194,030
Inpatient cost, 2004 $
Inpatient cost within the first 100 days96,26471,090-150,293
Total inpatient cost of the first year110,08277,057-175,348
Outpatient cost, 2004 $
Outpatient costs within the first 100 days4,589403-10,476
Total outpatient costs of the first year10,493558-23,397
Hospital stay, days
Hospital stay for the first 100 days3628-47
Total hospital stay for the first year3930-54
  • View full-size image.
  • Figure 1 

    Characteristics of costs during the first year. (A) Proportion of costs by phase after transplantation. Abbreviation: early phase, initial admission through 100 days after transplantation; later phase, from 101 days to 1 year (365 days) after transplantation. (B) Relationship between total costs and disease-free survival during the first year by each year of transplantation. Boxes indicate the median costs in 2004 dollars within the first year post-HDCT. Lines indicate the disease-free survival at 1 year. Numbers of patients in each transplant year are shown in the parentheses.

The changes in total costs during the first year post-HDCT by year of transplant were shown in Figure 1B. Both cost and days of hospital stay increased from 2000 to 2002, but both decreased from 2002 to 2004. One-year DFS for patients transplanted in each year improved steadily (46% in 2000 and 78% in 2004).

The results of the multivariate analyses of costs for the first year are shown in Table 4. For example, a ratio of 1.3 for costs corresponds to a 30% increase in costs (IC $35,042) when patients transplanted from unrelated donors are compared with those transplanted from related donors, controlling for all other variables (P < .01). Advanced risk diseases (IC $20,109, ratio 1.2, P = .04) were also associated with higher costs. Patients transplanted from female donors also had higher costs. However, there was an interaction with patient age such that the effect of donor sex was seen specifically in patients who were aged over 50 years (IC $38,960, ratio 1.5, P = .02). In the full model of early costs, considering both the baseline variables and post-transplant events occurring within the first 100 days, the following were the independent factors associated with high costs: grade II to IV aGVHD (IC $46,414, ratio 1.3, P < .01); late neutrophil recovery or without engraftment (IC $48,789, ratio 1.3, P < .01); VOD (IC $53,009, ratio 1.3, P < .01); IP/DAH (IC $40,741, ratio 1.2, P < .01); severe neurological toxicity (IC $43,639, ratio 1.2, P < .01) and in-hospital death (IC $50,476, ratio 1.3, P < .01). Severe infection (IC $17,553, ratio 1.1, P = .06), severe renal/bladder toxicity (IC $26,775, ratio 1.1, P = .08), and severe cardiac toxicity (IC $33,256, ratio 1.2, P = .07) were also found to be marginal predictors of high costs. Relapse in the early phase was not associated with higher costs. In the full model limited to patients who survived over 100 days, higher later costs were associated with extensive cGVHD (IC $7,003, ratio 1.7, P < .01), relapse (IC $15,069, ratio 2.9, P < .01), and in-hospital death (IC $23,435, ratio 4.7, P < .01) after controlling for the baseline variables and post-transplant events. Use of unrelated donors remained marginally predictive of high costs in the early phase (IC $19,341, ratio 1.1, P = .09) and in the later phase (IC $6,244, ratio 1.6, P = .05) in the full model.

Table 4. Multivariate analyses of costs within the first year
Baseline model Full model (early phase) Full model (later phase)
Variables∗Incremental costsRatio (95% CI)P-valueIncremental costsRatio (95% CI)P-valueIncremental costsRatio (95% CI)P-value
GVHD prophylaxis
Low dose MTX vs Standard dose MTX-19,5730.9 (0.7-1.0)0.126,9711.0 (0.9-1.2)0.58-7,3470.6 (0.4-1.0)0.05
No MTX vs Standard dose MTX-14,2820.9 (0.7-1.1)0.34-3,2460.9 (0.8-1.1)0.551,5721.1 (0.6-2.0)0.70
TCD vs Non TCD-3,7341.0 (0.8-1.2)0.80-8,7531.0 (0.8-1.2)0.852,5041.2 (0.7-2.2)0.54
HLA matching
Mismatched vs Matched5,5611.0 (0.8-1.3)0.7011,2181.1 (0.9-1.2)0.422,0471.2 (0.6-2.1)0.62
Donor type
Unrelated vs Related35,0431.3 (1.1-1.5)<0.0119,3411.1 (1.0-1.2)0.096,2441.6 (1.0-2.5)0.05
Disease
ALL vs Myeloid-17,9950.9 (0.7-1.1)0.1714,7561.1 (0.9-1.2)0.25-6,7880.6 (0.4-1.0)0.06
NHL vs Myeloid-7,4740.9 (0.8-1.2)0.588,8731.0 (0.9-1.2)0.511,8541.1 (0.7-1.9)0.60
Disease Status
Advanced risk vs Standard risk20,1091.2 (1.0-1.3)0.047,9701.0 (0.9-1.1)0.419071.1 (0.7-1.5)0.72
Donor sex
Female donor vs Male donor24,0271.2 (1.0-1.4)0.0416,4061.1 (1.0-1.2)0.142,1321.2 (0.7-1.8)0.50
Patient age
Aged 51 or older vs Aged 50 or younger1,0711.0 (0.9-1.2)0.9310,3961.0 (0.9-1.2)0.349171.1 (0.7-1.7)0.77
Stem cell source
Bone marrow vs Peripheral blood-1,6691.0 (0.8-1.2)0.89-11,9940.9 (0.8-1.1)0.35-3,9110.8 (0.5-1.2)0.24
CMV serological status
CMV positive vs CMV negative15,0151.1 (1.0-1.3)0.0811,9291.1 (1.0-1.2)0.171,2551.1 (0.8-1.5)0.59
Year of transplant
2000 vs 2004-13,0550.9 (0.6-1.3)0.59-47,2440.8 (0.6-1.0)0.06-19,4570.3 (0.1-0.7)<0.01
2001 vs 200412,2921.1 (0.8-1.6)0.60-24,5990.9 (0.7-1.1)0.30-4,7070.7 (0.3-1.7)0.45
2002 vs 200430,2311.3 (0.9-1.7)0.1519,0881.1 (0.9-1.4)0.38-5,2720.7 (0.3-1.5)0.32
2003 vs 200422,1241.2 (0.9-1.6)0.285131.0 (0.8-1.2)0.981,5931.1 (0.5-2.4)0.76
Recovery of neutrophil
Late or without recovery vs Early recovery 48,7891.3 (1.1-1.4)<0.01---
VOD
Yes vs No 53,0091.3 (1.1-1.5)<0.01---
IP/DAH
Yes vs No 40,7411.2 (1.1-1.4)<0.017,0061.7 (0.7-4.0)0.26
Infection
Yes vs No 17,5531.1 (1.0-1.2)0.064,1861.4 (0.9-2.0)0.14
Renal/bladder toxicity
Yes vs No 26,7751.1 (1.0-1.3)0.085,2421.5 (0.8-2.6)0.19
Neurological toxicity
Yes vs No 43,6391.2 (1.1-1.4)<0.016,9891.7 (0.9-3.2)0.13
Cardiac toxicity
Yes vs No 33,2561.2 (1.0-1.4)0.075,5661.5 (0.8-2.8)0.24
Grade II to IV acute GVHD
Yes vs No 46,4141.3 (1.1-1.4)<0.01---
extensive chronic GVHD
Yes vs No ---7,0031.7 (1.2-2.3)<0.01
Relapse
Yes vs No 17,8901.1 (0.9-1.3)0.2615,0692.9 (1.7-4.9)<0.01
In-hospital death
Yes vs No 50,4761.3 (1.1-1.5)<0.0123,4354.7 (2.7-8.1)<0.01

CI indicates confidence interval; early phase, initial admission through 100 days after transplantation; later phase, from 101 days to 1 year (365 days) after transplantation; GVHD, graft-versus-host disease; MTX, methotrexate; vs, versus; TCD, T cell depleted; ALL, acute lymphoblastic leukemia; AML, acute myelogenous leukemia; MDS, myelodysplastic syndrome; CML, chronic myelogenous leukemia; NHL, non-Hodgkin lymphoma; myelogenous included AML, MDS, and CML; CMV, cytomegalovirus; CMV positive, CMV positive serology in either donor or patient; CMV negative, CMV negative serology both in patient and donor; late or without recovery, late or without neutrophil engraftment; early recovery, early neutrophil engraftment within 15 days after HCT; VOD, veno-occlusive disease; IP, idiopathic pneumonia; DAH, diffuse alveolar hemorrhage.

∗Reference groups: standard dose MTX, TCD, HLA matched, related, myeloid, standard risk, male donor, aged 50 years or younger, peripheral blood, CMV negative, patients transplanted in 2004, patients with early recovery, no VOD, no IP/DAH, no infection, no renal/bladder toxicity, no neurological toxicity, no grade II to IV acute GVHD, no extensive chronic GVHD, did not relapse, did not die in the hospital.

Given the different clinical outcomes between related and unrelated donor transplantations, we performed a stratified analysis of donor type and confirmed the same results. In-hospital death was associated with the diagnosis of VOD, but not other complications in the early phase. In the later phase, patients with severe neurological, renal/bladder, or cardiac toxicities were more likely to die in the hospital. Exclusion of in-hospital death as a potential predictor would increase the IC of complications by $1,716-$22,253, except for neurological toxicity where the IC would decrease by $299 in the early phase. Exclusion of in-hospital death as a predictor of later costs would change the ICs between -$6,265 and $11,570.

When the total number of complications was considered, cost increased $20,228 on average per complication. OS at 1 year decreased with each additional complication with a particularly steep drop-off if 3 or more complications occurred (Figure 2A). If no complications occurred, costs during the first year were $79,222, consisting of $74,044 in the early phase, plus $6,080 in the later phase, after controlling for all the baseline characteristics as well as number of complications. Figure 2B compares the costs of patients dying in the early phase (n=63), those dying in the later phase (n=70), and surviving patients (n=182) throughout the first year post-HDCT. Surviving patients were the least costly (P < .01), whereas costs of patients dying within the first year were similar regardless of whether death occurred in the early or later phase (P = .53). Sixty-one percent of deceased patients accrued costs over $150,000 per patient, whereas 26% of surviving patients had costs above that amount.

  • View full-size image.
  • Figure 2 

    Relationship between costs and post-transplant events or death. (A) Costs and overall survival during the first year after transplantation by number of complications. Boxes indicate the estimated mean costs in 2004 dollars for the first year after transplantation, which were adjusted for patient and transplant characteristics and the 95% confidence interval were shown by the full range of upper and lower whiskers. Lines indicate the overall survival at 1 year. Late neutrophil recovery, veno-occlusive disease, idiopathic pneumonia/diffuse alveolar hemorrhage, severe infection, severe renal/bladder toxicity (≥grade 3), severe neurological toxicity (≥grade 3), severe cardiac toxicity (≥grade 3), grade II to IV acute GVHD, extensive chronic GVHD, relapse, and in-hospital death during the first year were considered. Numbers of patients with each number of complications are shown in the parentheses. (B) Costs during the first year after transplantation for deceased or surviving patients.

Back to Article Outline

Discussion 

In accordance with the previous literature 19, 20, 32, 40, 41, our data showed that the costs within the first 100 days (median $102,574) made up a substantial portion of the total costs within the first year for HDCT. This pattern of cost accrual differs from solid organ transplantation such as lung transplants, where long-term costs tend to be higher than initial or early costs 42, 43, 44. Advances in stem cell transplant techniques, supportive care, and the recent change of primary locus of care for HDCT from inpatient to outpatient are anticipated to reduce costs especially for low-risk patients [45], hopefully without increasing long-term costs. Our data showed that although there was a trend towards increased outpatient costs since 2002, these costs still comprise a small proportion of total costs within the first year at our institution.

In contrast to the previous finding by Bennett et al [17], but in agreement with Griffith et al, good clinical outcomes do not necessarily translate into lower costs and fewer days of hospital stay in our study. During the period of this study, our institution studied “low dose MTX” or “sirolimus based, without MTX” regimens. Costs during this period rose coincident with these studies [46], but then later fell, whereas DFS improved steadily. This non-linear change in costs over time was still detected in the multivariate analysis. Once physicians and support staff gain additional experience with the new GVHD prophylaxis regimens, treatment approaches, supportive care, and/or patient selection criteria we speculate this may result in both better outcomes and reduced costs for the last two years of the study period, as demonstrated by Freeman et al [16].

One concern about cost analyses of procedures with appreciable mortality rates is low costs may be seen if patients either die very early or have uncomplicated courses, two very different clinical outcomes. Our data showed that deceased patients were more costly compared to surviving patients even when they died in the early phase post-HDCT.

Although we expected complications to lead to higher costs, it is difficult to determine estimates of these effects from the literature. In multivariate analysis considering only pre-HDCT factors (those known as a patient starts HDCT), use of an unrelated donor 12, 13, 14 and advanced risk diseases were the significant predictors of high costs. When both pre- and post- HDCT factors were included in the model, post-HDCT complications, such as grade II to IV aGVHD, late neutrophil recovery or without engraftment, VOD, IP/DAH, severe neurological toxicity, and in-hospital death were associated with higher costs of between $40,741 and $53,009 within the first 100 days. If patients experienced extensive cGVHD, relapse, and in-hospital death beyond 100 days, costs of between $7,003 and $23,435 were added in the later time period. Our study showed a positive correlation between the development of grade II to IV aGVHD and decreased relapse rate if patients survive longer. Despite high costs associated with some complications, patients may ultimately experience clinical benefit in the long term.

We found the estimated costs were $79,222 (n=15) for the first year if no complication occurred; this was lower than similar estimates of costs of “uncomplicated patients” who received transplants before 2000, reportedly $119,096 for the initial hospitalization[12] and $109,594 for 5 years post-transplant [41] (Table 5). (All dollar values were converted to 2004 US dollars using the consumer price index for more effective comparison). These findings suggest a possible trend of cost reduction for uncomplicated cases transplanted recently. On the other hand, compared to the published literature, costs for infection or grade II to IV aGVHD have not changed significantly over time, but costs of VOD may be higher for patients who received transplant recently. Expensive supportive care may improve outcomes and increase cost.

Table 5. Costs of complications after transplantation
Cost of complication, mean (median)
PublicationType of transplantDiseaseTime horizonsuncomplicated transplant costsInfectionCMV infectionVODacute GVHDchronic GVHDMean (median) cost per patientMean (median) hospital days
Griffiths 1993[14]Allo-BMT '80-'87409 malignant or nonmalignant disease6 months (payer costs, ipt. costs only) +$101,939 +$88,573+$72,451$396,978 ($334,076)
Lee 2000[12]Allo-HCT '94-'97181 hematological malignancyinitial hospitalization$119,096+$20,224 +$24,586+$37,143 ($139,188)
Esperou 2004[13]Allo-HCT MCRCT '98-'0085 (RCT:200) hematological malignancy6 months 0-2 vs ≥3: +$17,010 +$5,358G0-1 vs ≥G2:+ $26,839+$32,952$112,659 ($102,727)initial hospitalization 34 (33)
Svahn 2006[41]Allo-HCT '98-'9993 malignant or nonmalignant disease, solid tumor5 years$109,594∗‡RR=1.33† RR=1.32†RR=1.35† initial hospitalization: ($54,967) 1y: ($126,065 ipt. + $16,366 opt.) 5y: ($142,347 ipt. + $27,738 opt. = $174,617 total)‡
Saito[25]Allo-HCT '00-'04315 hematological malignancyearly phase$74,044+$17,553 +$53,009+$46,414+$7,003($102,574)(36)
later phase$6,080 ($128,800)

All dollar values were converted to 2004 dollars for more effective comparison. CMV, cytomegalovirus; VOD, veno-occlusive disease; GVHD, graft-versus-host disease; Allo-, allogeneic; BMT, bone marrow transplantation; HCT, hematopoietic cell transplantation; RCT, randomized clinical trial; G, grade; RR, relative risk; d, days; y(s), year(s); ipt, inpatient; opt, outpatient; early phase, initial admission to 100 days after transplantation; later phase, from 101 days to 1 year (365 days) after transplantation.

∗Data were read from Figure 2 in the paper.

†Actual incremental costs over the baseline were not described in the paper. The ratios that appeared in the table were derived from a multivariate analysis of costs within the first year after transplantation.

‡1€=$1.3

Although results are derived from a single center study, the size of the study population with the same conditioning regimens and the 4-year period of study help mitigate concerns of generalizability. Our results can be used to project the financial implications of different complications, and to model effects of different approaches to prophylaxis. We could not capture costs accrued beyond the first year post-HDCT, however we believe longer follow-up is not necessary to estimate the cost implications of pre-transplant factors and early complications [41]. Most economic studies analyze costs because costs are true measures of resources used. In contrast, charges are usually multiple of costs and are driven by many local factors varying from institution to institution. Charges may be computed from costs using departmental ratios of costs to charges in each institution.

In conclusion, we found that costs for the first 100 days were the major cost contributors during the first year after transplantation, and inpatient costs accounted for the majority of the total costs. Use of unrelated donors and advanced disease status predict higher costs when considering only pre-HDCT factors. When post-HDCT events are considered, severe complications appear to be major cost driver with an average incremental cost of $20,228. If complications can be prevented or be treated by less costly, but effective procedures, significant cost savings might be achieved, along with better clinical outcomes.

Back to Article Outline

Acknowledgments 

This study was supported by National Heart, Lung, and Blood Institute grant (# P01 HL070149) and a Marx Fellowship. We thank our colleagues at the DFCI/BWH and our patients for sharing their experiences with us. We also thank Qiheng Yang, Tarrah Kirkpatrick, Daniel J. Quinn, Robin R. Junkins, Denise Sullivan, Benjamin S. Parsons, Celeste Daye, and Mohammed Yousuf for their help in gathering clinical and cost information.

Back to Article Outline

References 

  1. Thomas ED, Lochte HL, Lu WC, et al. Intravenous infusion of bone marrow in patients receiving radiation and chemotherapy. N Engl J Med. 1957;257:491–496
  2. Mielcarek M, Martin PJ, Leisenring W, et al. Graft-versus-host disease after nonmyeloablative versus conventional hematopoietic stem cell transplantation. Blood. 2003;102:756–762
  3. Sparrelid E, Hagglund H, Remberger M, et al. Risk factors for septicemia during aplastic period after allogeneic bone marrow transplantation. Transplant Proc. 1995;27:3530
  4. Hebart H, Loffler J, Meisner C, et al. Early detection of aspergillus infection after allogeneic stem cell transplantation by polymerase chain reaction screening. J Infect Dis. 2000;181:1713–1719
  5. Ljungman P, Lore K, Aschan J, et al. Use of a semi-quantitative PCR for cytomegalovirus DNA as a basis for pre-emptive antiviral therapy in allogeneic bone marrow transplant patients. Bone Marrow Transplant. 1996;17:583–587
  6. Ringden O, Le Blanc K. Allogeneic hematopoietic stem cell transplantation: state of the art and new perspectives. Apmis. 2005;113:813–830
  7. Lee SJ. Outcomes research in hematopoietic cell transplantation. In:  Blume KG,  Forman SJ,  Appelbaum FR editor. Thomas's Hematopoietic Cell Transplantation. 3rd ed.. Malden, Mass:: Blackwell; 2004;p. 434–446
  8. American Cancer Society. Cancer Facts & Figures 2006. Available at http://www.cancer.org/docroot/STT/stt_0.asp Accessed March 21, 2007.
  9. Insurance and Transplant Coverage Information: National Marrow Donor Program. Available at http://www.marrow.org/PATIENT/Plan_for_Tx/Planning_for_Tx_Costs/Insurance_and_Transplant_Cover/index.html. Accessed August 15, 2007.
  10. Kollman C, Weis T, Switzer GE, et al. Non-HLA barriers to unrelated donor stem cell transplantation. Bone Marrow Transplant. 2001;27:581–587
  11. Kline RM, Meiman S, Tarantino MD, et al. A detailed analysis of charges for hematopoietic stem cell transplantation at a children's hospital. Bone Marrow Transplant. 1998;21:195–203
  12. Lee SJ, Klar N, Weeks JC, et al. Predicting costs of stem-cell transplantation. J Clin Oncol. 2000;18:64–71
  13. Esperou H, Brunot A, Roudot-Thoraval F, et al. Predicting the costs of allogeneic sibling stem-cell transplantation: results from a prospective, multicenter, French study. Transplantation. 2004;77:1854–1858
  14. Griffiths RI, Bass EB, Powe NR, et al. Factors influencing third party payer costs for allogeneic BMT. Bone Marrow Transplant. 1993;12:43–48
  15. Matsuo K, Hamajima N, Morishima Y, Harada M. Hospital capacity and post-transplant survival after allogeneic bone marrow transplantation: analysis of data from the Japan Society for Hematopoietic Cell Transplantation. Bone Marrow Transplant. 2000;26:1061–1067
  16. Freeman M, Vose J, Bennett C, et al. Costs of care associated with high-dose therapy and autologous transplantation for non-Hodgkin's lymphoma: results from the University of Nebraska Medical Center 1989 to 1995. Bone Marrow Transplant. 1999;24:679–684
  17. Bennett CL, Armitage JL, Armitage GO, et al. Costs of care and outcomes for high-dose therapy and autologous transplantation for lymphoid malignancies: results from the University of Nebraska 1987 through 1991. J Clin Oncol. 1995;13:969–973
  18. Masaoka T, Hiraoka A, Okamoto S, et al. Asian Pacific cooperative study of allogeneic bone marrow transplantation. Int J Hematol. 1999;70:190–192
  19. de Lissovoy G, Hurd D, Carter S, et al. Economic analysis of unrelated allogeneic bone marrow transplantation: results from the randomized clinical trial of T-cell depletion vs unmanipulated grafts for the prevention of graft-versus-host disease. Bone Marrow Transplant. 2005;36:539–546
  20. Lee SJ, Zahrieh D, Alyea EP, et al. Comparison of T-cell-depleted and non-T-cell-depleted unrelated donor transplantation for hematologic diseases: clinical outcomes, quality of life, and costs. Blood. 2002;100:2697–2702
  21. Nichols WG, Corey L, Gooley T, et al. Rising pp65 antigenemia during preemptive anticytomegalovirus therapy after allogeneic hematopoietic stem cell transplantation: risk factors, correlation with DNA load, and outcomes. Blood. 2001;97:867–874
  22. Reusser P, Einsele H, Lee J, et al. Randomized multicenter trial of foscarnet versus ganciclovir for preemptive therapy of cytomegalovirus infection after allogeneic stem cell transplantation. Blood. 2002;99:1159–1164
  23. Andstrom EE, Ringden O, Remberger M, et al. Safety and efficacy of liposomal amphotericin B in allogeneic bone marrow transplant recipients. Mycose.s. 1996;39:185–193
  24. Svahn BM, Ringden O, Remberger M. Treatment costs and survival in patients with grades III-IV acute graft-versus-host disease after allogenic hematopoietic stem cell transplantation during three decades. Transplantation. 2006;81:1600–1603
  25. Saito AM, Zahrieh D, Cutler C, et al. Lower costs associated with hematopoietic cell transplantation using reduced intensity vs high-dose regimens for hematological malignancy. Bone Marrow Transplant. 2007;in press
  26. Przepiorka D, Weisdorf D, Martin P, et al. 1994 Consensus Conference on Acute GVHD Grading. Bone Marrow Transplan. 1995;15:825–828
  27. Shulman HM, Sullivan KM, Weiden PL, et al. Chronic graft-versus-host syndrome in man. A long-term clinicopathologic study of 20 Seattle patients. Am J Med. 1980;69:204–217
  28. McDonald GB, Hinds MS, Fisher LD, et al. Veno-occlusive disease of the liver and multiorgan failure after bone marrow transplantation: a cohort study of 355 patients. Ann Intern Med. 1993;118:255–267
  29. Cooke KR, Kobzik L, Martin TR, et al. An experimental model of idiopathic pneumonia syndrome after bone marrow transplantation: I. The roles of minor H antigens and endotoxin. Blood. 1996;88:3230–3239
  30. Robbins RA, Linder J, Stahl MG, et al. Diffuse alveolar hemorrhage in autologous bone marrow transplant recipients. Am J Med. 1989;87:511–518
  31. Chao NJ, Duncan SR, Long GD, et al. Corticosteroid therapy for diffuse alveolar hemorrhage in autologous bone marrow transplant recipients. Ann Intern Med. 1991;114:145–146
  32. Mishra V, Vaaler S, Brinch L. A prospective cost evaluation related to allogeneic haemopoietic stem cell transplantation including pretransplant procedures, transplantation and 1 year follow-up procedures. Bone Marrow Transplant. 2001;28:1111–1116
  33. Hsiao WC, Braun P, Yntema D, et al. Estimating physicians' work for a resource-based relative-value scale. N Engl J Med. 1988;319:835–841
  34. Bureau of Labor Statistics. Consumer Price Index. Available at http://www.bls.gov. Accessed March 1, 2005.
  35. Gold MR, Siegel JE, Russell LB, et al. Cost-effectiveness in Health and Medicine. Oxford University Press: New York; 1996;
  36. Mantel N. Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep. 1966;50:163–170
  37. Alyea EP, Li S, Kim H, et al. Sirolimus, tacrolimus and reduced-dose methotrexate as graft versus host disease (GVHD) prophylaxis after non-myeloablative stem cell transplantation: low incidence of acute GVHD compared with tacrolimus/methotrexate or cyclosporine/prednisone. Blood. 2004;104:Abstract
  38. Antin JH, Kim HT, Cutler C, et al. Sirolimus, tacrolimus, and low-dose methotrexate for graft-versus-host disease prophylaxis in mismatched related donor or unrelated donor transplantation. Blood. 2003;102:1601–1605
  39. Cutler C, Kim HT, Hochberg E, et al. Sirolimus and tacrolimus without methotrexate as graft-versus-host disease prophylaxis after matched related donor peripheral blood stem cell transplantation. Biol Blood Marrow Transplant. 2004;10:328–336
  40. van Agthoven M, Groot MT, Verdonck LF, et al. Cost analysis of HLA-identical sibling and voluntary unrelated allogeneic bone marrow and peripheral blood stem cell transplantation in adults with acute myelocytic leukaemia or acute lymphoblastic leukaemia. Bone Marrow Transplant. 2002;30:243–251
  41. Svahn BM, Alvin O, Ringden O, et al. Costs of allogeneic hematopoietic stem cell transplantation. Transplantation. 2006;82:147–153
  42. Vasiliadis HM, Collet JP, Penrod JR, et al. A cost-effectiveness and cost-utility study of lung transplantation. J Heart Lung Transplant. 2005;24:1275–1283
  43. Ramsey SD, Patrick DL, Albert RK, et al. The cost-effectiveness of lung transplantation. A pilot study. University of Washington Medical Center Lung Transplant Study Group. Chest. 1995;108:1594–1601
  44. Anyanwu AC, McGuire A, Rogers CA, et al. An economic evaluation of lung transplantation. J Thorac Cardiovasc Surg. 2002;123:411–418discussion 418-420
  45. Rizzo JD, Vogelsang GB, Krumm S, et al. Outpatient-based bone marrow transplantation for hematologic malignancies: cost saving or cost shifting?. J Clin Oncol. 1999;17:2811–2818
  46. Cutler C, Henry NL, Magee C, et al. Sirolimus and thrombotic microangiopathy after allogeneic hematopoietic stem cell transplantation. Biol Blood Marrow Transplant. 2005;11:551–557

PII: S1083-8791(07)00538-1

doi:10.1016/j.bbmt.2007.10.010

Biology of Blood and Marrow Transplantation
Volume 14, Issue 2 , Pages 197-207, February 2008

Article Tools

* Image Usage & Resolution