Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Abramson, R. G.
Right arrow Articles by Stuart, K. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Abramson, R. G.
Right arrow Articles by Stuart, K. E.
(Radiology. 2000;216:485-491.)
© RSNA, 2000

Health Policy and Practice

Cost-effectiveness of Hepatic Arterial Chemoembolization for Colorectal Liver Metastases Refractory to Systemic Chemotherapy1

Richard G. Abramson, BA, Max P. Rosen, MD, MPH, Laura J. Perry, MD, David P. Brophy, MD, S. Lynn Raeburn, RN and Keith E. Stuart, MD

1 From the Harvard Medical School, Boston, Mass (R.G.A., M.P.R., L.J.P., D.P.B., K.E.S.) and the Department of Radiology, Beth Israel Deaconess Medical Center, 330 Brookline Ave, Boston, MA 02215 (M.P.R., L.J.P., D.P.B., S.L.R., K.E.S.). From the 1999 RSNA scientific assembly. Received August 23, 1999; revision requested October 26; revision received November 23; accepted December 21. Address correspondence to M.P.R. (e-mail: mrosen2@caregroup.harvard.edu).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
PURPOSE: To calculate the cost-effectiveness of hepatic arterial chemoembolization (HACE) for the treatment of colorectal liver metastases (CLM) over a range of survival benefits and to determine the survival benefit that HACE must confer to meet three thresholds of cost-effectiveness.

MATERIALS AND METHODS: A spreadsheet model was used to estimate the marginal direct cost of HACE compared with palliative care from a payer’s perspective. Medicare reimbursement amounts represented costs, while probabilities of reembolization and complications were obtained from records of patients who underwent HACE. Marginal cost-effectiveness was calculated from marginal direct cost by varying the survival benefit of HACE compared with palliative care from 0 to 24 months. Break-even analyses were conducted to determine the survival benefit at which the cost-effectiveness of HACE would decrease below three threshold values derived from a literature review.

RESULTS: The marginal cost-effectiveness of HACE compared with palliative care, given survival benefits of 3, 6, and 12 months, was $82,385, $41,193, and $21,045 per life-year (LY) gained, respectively. Cost-effectiveness thresholds of $20,000 (strict), $50,000 (moderate), and $100,000 (generous) per LY gained required survival benefits of 12.63, 4.94, and 2.47 months, respectively, more than the expected baseline.

CONCLUSION: The cost-effectiveness of HACE for the treatment of CLM varies considerably according to the anticipated survival benefit. Results of future randomized controlled trials must demonstrate a survival benefit of nearly 5 months for HACE to meet the moderate cost-effectiveness standard of $50,000 per LY gained.

Index terms: Cost-effectiveness • Hepatic arteries, chemotherapeutic embolization, 952.1264, 953.1266 • Interventional procedures, comparative studies • Interventional procedures, technology, 952.1264, 953.1266 • Liver neoplasms, metastases, 761.3327


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Hepatic arterial chemoembolization (HACE) has emerged as a therapeutic and palliative option for a variety of hepatic lesions, including primary hepatocellular carcinoma (1–10), metastases from neuroendocrine tumors (11–14), and metastases from colorectal tumors (15–20). The technique, which involves selective cannulation of the hepatic artery and delivery of a chemotherapeutic agent in an embolic slurry, was developed from the observation that hepatic neoplasms derive their blood supply from the hepatic artery, whereas normal hepatic parenchyma derives its blood supply from the portal vein (21). Although phase 2 trials have generated copious data on the safety and effectiveness of HACE, to our knowledge, published studies of cost-effectiveness do not exist, in part because few randomized controlled trials have produced data on the survival benefit, as compared with palliative care (22). Information on cost-effectiveness will become more important as practitioners and payers seek to assess the future role of this technology in patient care.

The purpose of our study was to generate data on the cost-effectiveness of HACE for the treatment of colorectal liver metastases (CLM) that are refractory to systemic chemotherapy. To overcome the obstacle presented by the lack of substantive data on the survival benefit, we chose to calculate cost-effectiveness over a range of potential survival benefits. We also performed a break-even analysis to determine the survival benefit at which HACE would be considered to be cost-effective according to benchmarks in the literature.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Literature Review
A literature review (MEDLINE available at http://www.nlm.nih.gov/databases/freemedl.html) was performed to collect data on (a) studies of the cost-effectiveness of HACE, (b) the survival benefit of HACE compared with palliative care, and (c) the cost-effectiveness benchmarks for other medical technologies.

Patient Series
We compiled clinical data from a series of 21 patients who underwent HACE at our institution from April 1996 through December 1998. All patients had a diagnosis of colorectal carcinoma that was metastatic to the liver, as confirmed at surgery and/or biopsy, and that had been deemed to be unresponsive to systemic chemotherapy. After risks and benefits were explained and informed consent was obtained, each patient underwent HACE with 5–fluorouracil, mitomycin, and ethiodized oil followed by embolization with use of absorbable gelatin sponge (Gelfoam; Pharmacia & Upjohn, Peapack, NJ) (19). Complications were noted and treated appropriately. Patients were eligible for reembolization if the treated hepatic lesions grew and/or if new lesions developed.

For patients with metastatic lesions confined to one hepatic lobe, a single embolization procedure was planned. All patients were eligible for reembolization if the original lesion or lesions grew or if new lesions developed. The only contraindication for reembolization was portal venous thrombosis. In patients with metastatic lesions in both the right and left hepatic lobes, the lobe with the greatest metastatic burden was embolized first. If the portal vein remained patent, lesions in the alternate lobe were embolized during a second session. Specific lesions were considered for reembolization only if they failed to respond to the initial treatment. Survival times were noted from the date of first chemoembolization.

Calculation of Direct Costs
Overview of the model.—A computer spreadsheet (EXCEL 97; Microsoft, Seattle, Wash) was constructed to calculate the marginal direct cost of HACE compared with palliative care. Marginal direct cost was modeled as the total direct costs of HACE minus the total direct costs of palliative care (Fig 1). We considered only direct costs to the payer. Direct costs to the patient and indirect costs to society (eg, cost of lost labor) were excluded.



View larger version (20K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Diagram illustrates the equation used to calculate the marginal direct cost of HACE compared with palliative care.

 
Total direct costs of HACE included costs associated with the initial clinical evaluation, chemoembolization procedures, follow-up evaluation, and symptom control. The initial evaluation was modeled to include a physician office visit, a nonenhanced CT examination of the abdomen, and a laboratory measurement of serum carcinoembryonic antigen (CEA) level. The chemoembolization procedure itself was modeled to include both an inpatient hospitalization component and a practitioner procedural component. The inpatient hospitalization component incorporated all laboratory work performed at admission (eg, complete blood count, coagulation studies, measurement of creatinine level, and liver function tests) and postprocedural inpatient medications (eg, antiemetics and antibiotics). The cost of either inpatient hospitalization or the practitioner procedural component could increase due to complications (see later). Follow-up evaluation was modeled to include a physician office visit, nonenhanced abdominal CT examination, and measurement of CEA level; all of these occurred at programmed intervals. Symptom control was modeled to include separate components for iatrogenic (ie, postchemoembolization) and disease-related symptom relief. Our model included the simplifying assumption that patients who live longer due to HACE incur no additional costs for other diseases that might arise (or for coexistent morbid conditions that might continue) during the period of increased survival.

Total direct costs of palliative care included costs associated with the initial evaluation, follow-up evaluation, and symptom control. As with HACE, the initial evaluation was modeled to include a physician office visit, nonenhanced abdominal CT examination, and measurement of CEA level. Follow-up evaluation was modeled to include the same items at programmed intervals. Symptom control was modeled to include only disease-related symptom relief.

Data sources.—As detailed in Table 1, all input parameters derived from one of the following three sources: our internal data series, abstraction from referenced sources, or reasonable assumption. Direct costs to the payer were calculated by using 1998 Medicare reimbursement (24) as a proxy for inpatient hospitalization, procedural, and outpatient visit costs. Health Care Financing Administration (HCFA) Federal Upper Limit prices in 1998 (25) were used for outpatient drug costs; if HCFA prices were unavailable, average wholesale prices (25) for generic equivalents were substituted. Probabilities of reembolization were estimated by using the mean from our patient series. Schedules for follow-up visits and dosing regimens for analgesic medications were both modeled on assumption.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Parameters for Calculation of Marginal Direct Cost of HACE versus Palliative Care
 
Baseline survival expectation.—Our model required an estimate of baseline survival expectation (ie, survival for patients receiving only palliative care) because survival time affected the number of follow-up visits, which, in turn, affected total costs. We incorporated a baseline survival expectation of 12 months, which was based on Stangl et al’s (23) observation that patients with CLM who are treated with chemotherapy have a median survival time of 11.1 months. Since this figure overstates the survival time of patients who have already failed systemic chemotherapy, we also conducted sensitivity analyses with baseline survival expectations of 6 and 3 months (see later).

Treatment of complications.—Complications could increase the costs of chemoembolization in the following two ways: by changing the primary inpatient hospitalization diagnosis to a diagnosis-related group with higher weight and by adding practitioner procedural costs. In the calculation of inpatient hospitalization costs, the model included only major complications, as minor complications did not alter the inpatient hospitalization diagnosis-related groups. In the calculation of practitioner procedural costs, the model included extra procedures (eg, esophagogastroduodenoscopy) as warranted. Probabilities of all complications and additional procedures were estimated by using the mean from our patient series.

Calculation of Cost-effectiveness
The marginal cost-effectiveness of HACE was modeled as the marginal direct cost of HACE divided by the survival benefit of HACE. (Survival benefit was defined as additional survival beyond the expected baseline survival for patients receiving only palliative care.) To calculate cost-effectiveness over a range of potential survival benefits, the denominator was maintained as an independent variable and varied from 0 to 24 months. Discounting was not performed due to the short time horizon.

Break-even Analysis
A break-even analysis was performed to determine the survival benefit at which HACE would be considered to be cost-effective according to benchmarks obtained from our literature review.

Sensitivity Analyses
Sensitivity analyses were conducted to test the robustness of the model to variation in all key input parameters, including the cost of nonenhanced CT examination, cost of measuring the CEA level, cost of chemoembolization (both inpatient hospitalization and practitioner procedural components), complication rates, mean number of reembolization procedures per patient, cost of postchemoembolization symptom-control regimen, and baseline survival expectation.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Literature Review
Studies on the cost-effectiveness of HACE.—Our literature review yielded no studies on the cost-effectiveness of HACE.

Data on the survival benefit of HACE compared with palliative care.—We found no randomized controlled trial of HACE for the treatment of CLM. (We did, however, find three randomized controlled trials of HACE for the treatment of unresectable hepatocellular carcinoma [4,8,9]; all three studies revealed no statistically significant difference in survival between patients who underwent HACE and patients who received only palliative care.)

Cost-effectiveness benchmarks.—Our search for cost-effectiveness benchmarks yielded an array of cost-effectiveness estimates for other medical interventions, including cancer treatments (Table 2) (26,27). These estimates were derived from findings of studies with nonuniform analytic approaches (quality-adjusted versus non–quality-adjusted, perspective of the payer versus perspective of society, etc). On the basis of these benchmarks and other recommendations (28), we selected $20,000, $50,000, and $100,000 per life-year (LY) gained as strict, moderate, and generous thresholds, respectively, for cost-effective care.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Selected Cost-effectiveness Benchmarks for Other Medical Interventions
 
Patient Series
Of the 21 patients in our series, seven died, and 14 were still alive as of March 1999. The seven patients who died each underwent a mean of 1.57 additional reembolization procedure. Survival times from the first chemoembolization procedure ranged from 83 to 1,097 days (2.8–36.6 months; mean, 17.4 months; median, 21.1 months).

Calculation of Direct Costs
As shown in Table 3, our computer model calculated a marginal direct cost of HACE compared with palliative care of $20,596–$21,494, depending on the survival benefit. Total direct costs of HACE increased slightly with increased survival benefit because of more follow-up visits.


View this table:
[in this window]
[in a new window]

  		TABLE 3. Marginal Cost and Cost-effectiveness of HACE versus Palliative Care as a Function of Survival Benefit
 
Calculation of Cost-effectiveness
Table 3 demonstrates that as survival benefit varied from 0 to 24 months, the marginal cost-effectiveness of HACE decreased hyperbolically. For survival benefits of 3, 6, and 12 months, the marginal cost-effectiveness of HACE was $82,385, $41,193, and $21,045 per LY gained, respectively. Figure 2 depicts the relationship between marginal cost-effectiveness and survival benefit.



View larger version (63K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Line graph depicts the marginal cost-effectiveness of HACE compared with palliative care as a function of survival benefit.

 
Break-even Analysis
According to our model, a strict cost-effectiveness threshold of $20,000 per LY gained required a survival benefit of 12.63 months more than the expected baseline survival. A moderate threshold of $50,000 per LY gained required a survival benefit of 4.94 months. A generous threshold of $100,000 per LY gained required a survival benefit of 2.47 months. Arrows in Figure 2 indicate these values.

Sensitivity Analyses
Table 4 demonstrates the robustness of our model to variation in key input parameters. The results of the model did not change substantially with 25% increases or decreases in the cost of nonenhanced CT examination and the cost of measuring the CEA level. The model was similarly robust to the substitution of complication rates derived from Berger et al’s (29) and Chung et al’s (30) reviews of chemoembolization complications. (The substitutions for probability were as follows: pulmonary embolism, .006; sepsis and/or pneumonia, .01; complication requiring laparotomy, 0; complication requiring a ventilation-perfusion scan, .006; complication requiring percutaneous drainage, .02; complication requiring laparotomy, 0; and complication requiring esophagogastroduodenoscopy, .01). Neither an increase or decrease in the postchemoembolization medication regimen nor a decrease the baseline survival expectation (ie, survival for patients receiving palliative care only) to 6 and 3 months substantially alter the results of the model.


View this table:
[in this window]
[in a new window]

  		TABLE 4. Sensitivity Analyses
 
The results of the model were somewhat influenced by changes in the cost of chemoembolization, the mean number of reembolization procedures per patient, and the schedule of follow-up visits after HACE. A decrease in the marginal cost of HACE compared with palliative care (with an associated decrease in the survival benefit required to meet all three cost-effectiveness thresholds) was observed after decreases were made in the cost of chemoembolization (the decrease in hospital component was more than the decrease in the practitioner procedural component), the mean number of reembolization procedures per patient, and the frequency of follow-up visits after HACE. An increase in the marginal cost of HACE compared with palliative care (with an associated increase in the survival benefit required to meet all cost-effectiveness thresholds) was observed after increases were made in the cost of chemoembolization (the increase in the hospital component was more than the increase in the practitioner procedural component), the mean number of reembolization procedures per patient, and the frequency of follow-up visits after HACE.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
 
Our objective was to generate data on the cost-effectiveness of HACE for the treatment of CLM that is refractory to systemic chemotherapy. Since there are no substantive data on the survival benefit of HACE compared with palliative care, we calculated its cost-effectiveness across a range of potential survival benefits. Note that although we report the survival data of the patients in our series, these data were not used in our cost-effectiveness model, which was designed to vary the survival benefit as an independent variable from zero to 24 months.

An alternative approach would have been to estimate the survival benefit of HACE by comparing data from phase 2 trials of HACE with the data of observational studies on untreated CLM (ie, palliative care). We rejected this approach primarily because it would have left our intervention and control groups unmatched for the criteria that affect survival, including severity and progression of illness. Survival after the diagnosis of CLM varies considerably rate from a mean of 7 months to a maximum of 3–5 years (23,31) according to tumor growth, so an estimate of the survival benefit of HACE that is based on different patients from different studies would have been virtually meaningless.

Aside from this important theoretic concern, practical limitations to this approach included the lack of observational studies of patients with CLM refractory to systemic chemotherapy, as well as the fact that observational studies and phase 2 trials tend to begin their survival clocks at different times; observational studies begin from the initial diagnosis of hepatic metastases, and phase 2 trials begin from the date of first chemoembolization. In short, we agree with other observers (22) that randomized controlled trials must be performed to obtain useful survival data, and, in the absence of such studies, we believe that our approach was the most methodologically sound approach.

Our results demonstrate that the cost-effectiveness of HACE varies considerably with expected survival benefit (Table 3, Fig 2). For survival benefits of 3, 6, and 12 months, the marginal cost-effectiveness of HACE compared with palliative care was $82,385, $41,193, and $21,045 per LY, respectively. Cost-effectiveness thresholds of $20,000 (strict), $50,000 (moderate), and $100,000 (generous) per LY require survival benefits of 12.63, 4.94, and 2.47 months, respectively, more than the expected baseline survival. These data were based on direct costs to the payer only. It is unlikely, however, that our results would have changed substantially had our analysis been repeated from the perspective of society, since both direct costs to the patient and indirect costs to society would have been minimal in this population of nonworking patients with short life expectancies.

Our results contribute directly to the technologic assessment of HACE by providing cost-effectiveness data to supplement existing information on the safety and effectiveness of HACE. Once future randomized controlled trials yield more definitive data on the survival benefit of HACE, the true cost-effectiveness of HACE will emerge readily from the data in Table 3. Meanwhile, practitioners and payers can generate preliminary cost-effectiveness estimates by using their own empirical assessments of the survival benefit of HACE. This exercise will be important in the setting of variation between local institutions in HACE outcomes due to either differences in the severity of patients’ disease or to different levels of provider training, experience, and skill. Our data may be used to craft arguments for regional centers of excellence that would perform HACE, with higher survival benefits and cost-effectiveness.

An important implication follows from the observation that a moderate cost-effectiveness standard of $50,000 per LY gained requires a survival benefit of nearly 5 months more than the baseline. If future randomized controlled trials fail to demonstrate a survival benefit of this magnitude, our results would cast doubt on the use of HACE for the treatment of CLM refractory to systemic chemotherapy, especially in light of recent questions regarding the safety and effectiveness of HACE (22).

Any such conclusions must be drawn with an understanding of our study limitations, the most important of which is the fact that our cost-effectiveness data were not adjusted for quality of life. It is possible to argue, then, that even if the true survival benefit of HACE is low, the benefit of the procedure to the patient remains high because of its effect on the palliation of symptoms. While we lack the quantitative data with which to dispute this claim, we note the concerns of other clinicians who have suggested that HACE, by virtue of its associated morbidity and potential complications (29,30,32), may not improve quality of life and may, in fact, harm it (22). Our present study findings lay the groundwork for future investigations of the quality-adjusted cost-effectiveness of HACE. Such studies might include the use of a Markov chain approach to model transitions of patients between different quality-adjusted health states.

Additional limitations are derived from our data sources and model design. Although our use of Medicare reimbursement data is an accepted substitute for detailed resource-based cost accounting (33), our estimate of direct costs would have been more accurate had we used the latter method (34). Our simplifying assumption that patients who undergo HACE incur no additional costs due to other diseases during their period of increased survival implies a minor underestimation of the marginal direct costs of HACE. We did not apply a discount rate (to adjust for the effect of price inflation on future costs) because of the short time horizon (35); if we had, the net present value of all cost calculations would have decreased slightly.

Finally, sensitivity analyses demonstrated that our model was robust to changes in the cost of nonenhanced CT examination, cost of measuring the CEA level, complication rates, and baseline survival assumption, but it demonstrated that the model was less robust to changes in the cost of chemoembolization, reembolization rate, and follow-up regimen after HACE. Our results, therefore, become less applicable in settings in which costs differ greatly from our Medicare-derived estimates, in which reembolization rates are substantially higher or lower, and in which follow-up visit schedules are markedly different. It should be reiterated that our chemoembolization protocol is only one of several protocols currently in use (22).

A final note concerns the technical and theoretic issues surrounding the application of cost-effectiveness data in general. From a technical standpoint, cost-effectiveness data on a particular medical technology may be compared with either benchmarks from other interventions or absolute cost-effectiveness standards. The dangers of weighing cost-effectiveness estimates for different interventions have been well documented (36); meaningful comparison requires homogeneity among the various source studies with regard to the scope of analysis, time horizon, assignment of utility values, and various statistical adjustments (eg, inflation, future discounting, international currencies). Meanwhile, the use of absolute standards suffers from the lack of consistent thresholds for cost-effectiveness. Despite our attempt to cull reasonable cost-effectiveness thresholds from the literature, our choices of $20,000, $50,000, and $100,000 per LY gained were still somewhat arbitrary, since many of the benchmarks listed in Table 2 were derived from analyses that were adjusted for quality of life or that were conducted from the perspective of society. We would caution the reader against overzealous comparison of our data with cost-effectiveness benchmarks from methodologically noncompatible studies.

From a theoretic perspective, cost-effectiveness analysis is most useful when there is a consensus regarding its underlying premise, that is, a consensus that finite health care resources should be allocated to maximize the benefit per unit cost. Uncertainty regarding this fundamental paradigm arises at both the population and doctor-patient levels. At the population level, resource allocation by means of cost-benefit analysis is better accepted in a public program or in an individual health plan (in which the payer is directly accountable to beneficiaries for the provision of fair access to care) than among members of the general public of whom many individuals still accept the ability to pay as a legitimate determinant of access to resources above a certain minimal benefits standard (37). At the doctor-patient level, cognitive dissonance stems from conflicting goals and competing notions of fairness. Patients express dissatisfaction with health care costs, yet they reject arguments that resources should be withheld because the resources might provide greater benefit elsewhere. Practitioners, meanwhile, are bound by the Hippocratic oath to do everything to preserve and enhance life, yet they seek a rational basis for the allocation of care when they assume the financial risk for patients. Progress on this issue will require a continual dialogue about how to best use the results from cost-effectiveness studies.

In conclusion, we envision that these study findings will encourage future work along at least three different fronts. First, the technologic assessment effort would benefit considerably from randomized controlled trials of HACE for the treatment of CLM. Second, our analysis sets the stage for investigations of the quality-adjusted cost-effectiveness of HACE. Third, we eagerly await further dialogue regarding the underlying assumptions, proper venues, and appropriate applications of cost-effectiveness analysis.


   FOOTNOTES
 
Abbreviations: CEA = carcinoembryonic antigen, CLM = colorectal liver metastases, HACE = hepatic arterial chemoembolization, LY = life year

Author contributions: Guarantors of integrity of entire study, R.G.A., M.P.R.; study concepts and design, R.G.A., M.P.R.; definition of intellectual content, R.G.A., M.P.R.; literature research, R.G.A.; clinical studies, M.P.R., L.J.P., D.P.B., S.L.R., K.E.S.; data acquisition, all authors; data analysis, R.G.A.; statistical analysis, R.G.A.; manuscript preparation, R.G.A.; manuscript editing, R.G.A., M.P.R., L.J.P., D.P.B., K.E.S.; manuscript review, M.P.R., L.J.P., D.P.B., K.E.S.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 

  1. Yamada R, Sato M, Kawabata M, Nakatsuka H, Nakamura K, Takashima S. Hepatic artery embolization in 120 patients with unresectable hepatoma. Radiology 1983; 148:397-401.[Abstract/Free Full Text]
  2. Sakurai M, Okamura J, Kuroda C. Transcatheter chemo-embolization effective for treating hepatocellular carcinoma: a histopathologic study. Cancer 1984; 54:387-392.[Medline]
  3. Venook AP, Stagg RJ, Lewis BJ, et al. Chemoembolization for hepatocellular carcinoma. J Clin Oncol 1990; 8:1108-1114.[Abstract]
  4. Pelletier G, Roche A, Ink O, et al. A randomized trial of hepatic arterial chemoembolization in patients with unresectable hepatocellular carcinoma. J Hepatol 1990; 11:181-184.[Medline]
  5. Bismuth H, Morino M, Sherlock D, et al. Primary treatment of hepatocellular carcinoma by arterial chemoembolization. Am J Surg 1992; 163:387-394.[Medline]
  6. Stuart K, Stokes K, Jenkins R, Trey C, Clouse M. Treatment of hepatocellular carcinoma using doxorubicin/ethiodized oil/gelatin powder combination. Cancer 1993; 72:3202-3209.[Medline]
  7. Kanematsu T, Matsumata T, Shirabe K, et al. A comparative study of hepatic resection and transcatheter arterial embolization for the treatment of primary hepatocellular carcinoma. Cancer 1993; 71:2181-2186.[Medline]
  8. Madden MV, Krige JEJ, Bailey S, et al. Randomised trial of targeted chemotherapy with lipiodol and 5-epidoxorubicin compared with symptomatic treatment for hepatoma. Gut 1993; 34:1598-1600.[Abstract/Free Full Text]
  9. Group d’Etude et de Traitement du Carcinome Hépatocellulaire. A comparison of lipiodol chemoembolization and conservative treatment for unresectable hepatocellular carcinoma. N Engl J Med 1995; 332:1256-1261.[Abstract/Free Full Text]
  10. Solomon B, Soulen MC, Baum RA, Haskal ZJ, Shlansky-Goldberg RD, Cope C. Chemoembolization of hepatocellular carcinoma with cisplatin, doxorubicin, mitomycin-C, ethiodol, and polyvinyl alcohol: prospective evaluation of response and survival in a U.S. population. J Vasc Interv Radiol 1999; 10:793-798.
  11. Hajarizadeh H, Ivancev K, Mueller CR, Fletcher WS, Woltering EA. Effective palliative treatment of metastatic carcinoid tumors with intra-arterial chemotherapy/chemoembolization combined with octreotide acetate. Am J Surg 1992; 163:479-483.[Medline]
  12. Therasse E, Breittmayer F, Roche A, et al. Transcatheter chemoembolization of progressive carcinoid liver metastasis. Radiology 1993; 189:541-547.[Abstract/Free Full Text]
  13. Ruszniewski P, Rougier P, Roche A, et al. Hepatic arterial chemoembolization in patients with liver metastases of endocrine tumors: a prospective phase II study in 24 patients. Cancer 1993; 71:2624-2630.[Medline]
  14. Drougas JG, Anthony LB, Blair TK, et al. Hepatic artery chemoembolization for management of patients with advanced metastatic carcinoid tumors. Am J Surg 1998; 175:408-412.[Medline]
  15. Taniguchi H, Takahashi T, Yamaguchi T, Sawai K. Intraarterial infusion chemotherapy for metastatic liver tumors using multiple anti-cancer agents suspended in a lipid contrast medium. Cancer 1989; 64:2001-2006.[Medline]
  16. Lang EK, Brown CL. Colorectal metastases to the liver: selective chemoembolization. Radiology 1993; 189:417-422.[Abstract/Free Full Text]
  17. Kameyama M, Imaoka S, Fukuda I, et al. Delayed washout of intratumor blood flow is associated with good response to intraarterial chemoembolization for liver metastasis of colorectal cancer. Surgery 1993; 114:97-101.[Medline]
  18. Martinelli DJ, Wadler S, Bakal CW, et al. Utility of embolization or chemoembolization as second-line treatment in patients with advanced or recurrent colorectal carcinoma. Cancer 1994; 74:1706-1712.[Medline]
  19. Sanz-Altamira PM, Spence LD, Huberman MS, et al. Selective chemoembolization in the management of hepatic metastases in refractory colorectal carcinoma: a phase II trial. Dis Colon Rectum 1997; 40:770-775.[Medline]
  20. Tellez C, Benson AB, III, Lyster MT, et al. Phase II trial of chemoembolization for the treatment of metastatic colorectal carcinoma to the liver and review of the literature. Cancer 1998; 82:1250-1259.[Medline]
  21. Breedis C, Young G. The blood supply of neoplasms in the liver. Am J Pathol 1954; 30:969-985.
  22. Miller DL. First, do no harm. Radiology 1996; 198:10-12.[Free Full Text]
  23. Stangl R, Altendorf-Hofmann A, Charnley RM, Scheele J. Factors influencing the natural history of colorectal liver metastases. Lancet 1994; 343:1405-1410.[Medline]
  24. National Heritage Insurance Company. Medicare B bulletin. Health Care Financing Administration Web site.; Available at: http://www.hcfa.gov. Accessed March 1999..
  25. 1998 Drug Topics Red Book Montvale, NJ: Medical Economics, 1998.
  26. Smith TJ, Hillner BE, Desch CE. Efficacy and cost-effectiveness of cancer treatment: rational allocation of resources based on decision analysis. J Natl Cancer Inst 1993; 85:1460-1474.[Free Full Text]
  27. Smith TJ, Hillner BE. Decision analysis: a practical example. Oncology (Huntingt) 1995; 9(suppl 11):37-45.[Medline]
  28. Laupacis A, Feeny D, Detsky AS, Tugwell PX. How attractive does a new technology have to be to warrant adoption and utilization? Tentative guidelines for using clinical and economic evaluations. Can Med Assoc J 1992; 146:473-481.[Abstract]
  29. Berger DH, Carrasco CH, Hohn DC, Curley SA. Hepatic artery chemoembolization or embolization for primary and metastatic liver tumors: post-treatment management and complications. J Surg Oncol 1995; 60:116-121.[Medline]
  30. Chung JW, Park JH, Han JK, et al. Hepatic tumors: predisposing factors for complications of transcatheter oily chemoembolization. Radiology 1996; 198:33-40.[Abstract/Free Full Text]
  31. Burke D, Allen-Mersh TG. Colorectal liver metastases. Postgrad Med J 1996; 72:464-469.[Abstract]
  32. Borner M, Castiglione M, Triller J, et al. Considerable side effects of chemoembolization for colorectal carcinoma metastatic to the liver. Ann Oncol 1992; 3:113-115.[Abstract/Free Full Text]
  33. Neuman ML, Murphy BD, Rosen MP. Bedside placement of peripherally inserted central catheters: a cost-effectiveness analysis. Radiology 1998; 206:423-428.[Abstract/Free Full Text]
  34. Picus D. Comparing competing medical procedures: costs or charges—what should it matter?. Radiology 1996; 199:623-625.[Free Full Text]
  35. Udvarhelyi IS, Colditz GA, Rai A, Epstein AM. Cost-effectiveness and cost-benefit analysis in the medical literature: are the methods being used correctly?. Ann Intern Med 1992; 116:238-244.
  36. Mason J, Drummond M, Torrance G. Some guidelines on the use of cost effectiveness league tables. BMJ 1993; 306:570-572.
  37. Blendon RJ, Brodie M, Hyams TS, Benson JM. The American public and the critical choices for health system reform. JAMA 1994; 271:1539-1544.[Medline]



This article has been cited by other articles:


Home page
 J Natl Cancer Inst MonogrHome page
D. Provenzale and R. N. Gray
Colorectal Cancer Screening and Treatment: Review of Outcomes Research
J Natl Cancer Inst Monographs, October 1, 2004; 2004(33): 45 - 55.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Abramson, R. G.
Right arrow Articles by Stuart, K. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Abramson, R. G.
Right arrow Articles by Stuart, K. E.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE