Cost-effectiveness of Percutaneous Radiofrequency Ablation for Malignant Hepatic Neoplasms

Cost-effectiveness of Percutaneous Radiofrequency Ablation for Malignant Hepatic Neoplasms Sanjay K. Shetty, MD,1 Max P. Rosen, MD, MPH, Vassilios Raptopoulos, MD, and S. Nahum Goldberg, MD

PURPOSE: Percutaneous radiofrequency (RF) ablation is a promising technique for the treatment of hepatic malignancies. However, its cost-effectiveness has not been established. The purpose of this study is to determine the cost-effectiveness of RF ablation compared to palliative care in the treatment of hepatocellular cancer and colorectal liver metastases. This study also seeks to evaluate the effects of transition from traditional to newly implemented prospective outpatient reimbursement mechanisms on RF ablation cost-effectiveness. MATERIALS AND METHODS: The marginal direct costs of a percutaneous RF ablation treatment strategy were compared to palliative care over a range of survival benefits with use of a cost-effectiveness model built from the perspective of the payer. Variables used in the model, including complication rates and procedure efficacy, were obtained from the literature and the authors’ experience with 46 consecutive patients. RESULTS: The cost-effectiveness of a standardized percutaneous RF ablation treatment strategy compared to palliative care was $20,424, $11,407, $5,034, and $3,492, respectively, per life-year (LY) gained when marginal median survival conferred by RF ablation is 6 months, 1 year, 3 years, and 5 years. The RF ablation treatment strategy would be required to generate 6.14, 2.26, and 1.10 months of marginal median survival benefit to achieve strict ($20,000/LY gained), moderate ($50,000/LY gained), and generous ($100,000/LY gained) cost-effectiveness thresholds. Cost-effectiveness was sensitive to the number of lifetime treatments, hours of observation time, frequency of follow-up evaluations, cost of abdominal computed tomography, and decision to perform RF ablation as an inpatient or outpatient. CONCLUSION: Percutaneous RF ablation is a cost-effective treatment strategy compared to palliative care and has likely already achieved the survival benefit required to meet even a strict cost-effectiveness criterion. Dependence on reimbursement mechanism highlights the importance of concordance between policy and RF ablation technology. The results of this study allow flexible application of cost-effectiveness data despite current uncertainties in treatment and survival data and heterogeneity in treatment populations. Index terms:

Cost-effectiveness

•

Liver neoplasms, therapy

•

Radiofrequency (RF) ablation

J Vasc Interv Radiol 2001; 12:823– 833 Abbreviations: APC ⫽ Ambulatory Procedure Code, CLM ⫽ colorectal liver metastases, HCC ⫽ hepatocellular carcinoma, LY ⫽ life year, OPPS ⫽ Outpatient Prospective Payment System, RF ⫽ radiofrequency

RADIOFREQUENCY (RF) ablation is a promising technique for the treatment of malignant liver neoplasms. A growing From Harvard Medical School and Department of Radiology, Beth Israel Deaconess Medical Center, 330 Brookline Avenue, E/AN-248, Boston, Massachusetts 02215. Received December 14, 2000; revision requested February 5, 2001; revision received February 15; accepted February 16. From the 2001 SCVIR Annual Meeting. Address correspondence to M.P.R.; E-mail: [email protected] 1 Current address: Department of Internal Medicine, Lahey Clinic, Burlington, Massachusetts. © SCVIR, 2001

body of literature has documented encouraging results with percutaneous application of RF ablation for hepatocellular carcinoma (HCC) (1–5) and colorectal liver metastases (CLM) (1,2,5–8), as well as for other hepatic metastases (9). This minimally invasive treatment strategy is particularly appealing because only 10%– 20% of patients with either disease are appropriate candidates for surgical resection, the current gold standard therapy (10,11), because of the extent of tumor involvement or comorbid conditions. Percutaneous RF ablation involves the insertion

of one or more electrodes into the tumor nodule under imaging guidance. A highfrequency alternating electrical current is then applied, causing agitation of ions in the surrounding tissues and generation of frictional heat. Increased temperatures initiate protein denaturation and the development of coagulative necrosis in tissues surrounding the RF ablation probe (12,13). Several aspects of the procedure make it particularly appealing as a potential low-cost treatment strategy: RF ablation obviates the need for an open

823

824

•

July 2001

Cost-effectiveness of RF Ablation for Hepatic Neoplasms

JVIR

Table 1 Sensitivity Analysis: Alternate Scenarios of Reimbursement

Alternate Scenario

Cost/Treatment Session

Outpatient RFA (Current) Outpatient RFA (OPPS) Inpatient RFA

4,321.98 3,482.79 7,181.67

Cost-Effectiveness of RFA Compared to Palliative Care ($/LY Gained)*

Break-Even Analysis†

6 mo

1y

2y

3y

5y

$100,000/ LY Gained

$50,000/ LY Gained

$20,000/ LY Gained

20,424 17,755 29,518

11,407 10,072 15,954

6,731 6,064 9,004

5,034 4,589 6,550

3,492 3,225 4,402

1.10 0.94 1.66

2.26 1.93 3.41

6.14 5.23 9.25

* Marginal median survival with RFA. † Months marginal median survival with RFA compared to palliative care.

procedure, is easily repeated for treatment failures or new hepatic lesions, and can be performed on an outpatient basis. Although percutaneous RF ablation is frequently lauded as costeffective (3), this contention has never been formally tested. The large number of patients with HCC or CLM who are potential candidates for ablation demands that an ideal strategy be efficacious and cost-effective. In this study, we sought to evaluate the cost-effectiveness of percutaneous RF ablation compared to palliative care in the treatment of two well-studied forms of malignant liver neoplasm: HCC and CLM. Our model of the marginal direct costs incurred by a percutaneous RF ablation treatment strategy incorporates costs of the RF ablation treatment procedure(s) as well as a regimen of follow-up evaluations. Cost-effectiveness of the RF ablation treatment strategy was calculated over a range of potential survival benefits to allow cost-effectiveness analysis of percutaneous RF ablation in the absence of long-term survival data.

MATERIALS AND METHODS Overview This study was composed of a literature review, examination of our patient experience, creation of a cost-effectiveness model for percutaneous RF ablation compared to palliative care, determination of cost-effectiveness and break-even points, and sensitivity analyses. Literature Review A review of the literature was performed to collect data on (i) the natural history of HCC and CLM and (ii) the ef-

fectiveness of percutaneous RF ablation. These published data contributed to the selection of baseline assumptions incorporated into our model of RF ablation costeffectiveness. The search was performed with use of PaperChase (Boston, MA) and the following index terms: (i) hepatocellular carcinoma AND radiofrequency ablation, and (ii) colon AND carcinoma AND metastases AND radiofrequency ablation. In addition, the term “radiofrequency (RF) ablation” was used to search the index of the journal Radiology. Patient Series The records of 46 consecutive patients who presented to our institution for percutaneous RF ablation of liver neoplasms were reviewed. These patients were initially treated between October 1998 and August 2000 for treatment of either HCC or metastatic lesions. A total of 73 lesions were treated in 83 sessions. Mean lesion size in this series was 3.1 cm ⫾ 1.1. All patients selected for RF ablation were evaluated initially with contrastenhanced abdominal computed tomography (CT) or magnetic resonance (MR) imaging. Patients arrived at the hospital on the day of the procedure. The procedure was performed under either conscious sedation or general anesthesia at the discretion of the interventional radiologist and referring physician. Procedure guidance in all cases was completed with ultrasonography (US) and CT fluoroscopy to ensure adequate visualization of target lesions. RF ablation was performed with 17-gauge internally cooled electrodes in either single or cluster configurations (Cool-tip RF system; Radionics, Burlington, MA) (7,8) with use

of a previously described pulsed RF algorithm (14). Tumors were treated for 9 –12 minutes, with termination of internal cooling before electrode withdrawal to ensure adequate heating of tissues immediately adjacent to the electrode tip. Abdominal CT with and without intravenous contrast material was completed immediately after the procedure to assess treatment efficacy and allow immediate repeat RF ablation if necessary while the patient was still on the treatment table. Patients were placed under observation after the procedure and typically discharged the same afternoon or, occasionally, the following day. Patients were routinely evaluated with abdominal CT approximately 1 month after procedure and every 3 months thereafter, with repeat RF ablation treatments performed as necessary for incomplete treatment or metachronous lesions that became apparent only on later scans. Calculation of Costs: Summary of Model A computer spreadsheet (Excel 98; Microsoft, Seattle, WA) was created to evaluate the marginal direct costs of percutaneous RF ablation compared with a baseline of palliative care. Marginal direct cost was calculated as the total direct costs of RF ablation minus the total direct costs of palliative care. The analysis was conducted from the perspective of the payer (Medicare) and based on total reimbursement (Table 1). Total direct costs associated with RF ablation included three major components: initial evaluation, RF ablation procedure(s), and follow-up evalua-

Volume 12

Number 7

tions that were scheduled at varying intervals. The initial evaluation included an outpatient office visit, abdominal CT, and determination of a serum tumor marker. To account for different practice patterns, the direct cost of each RF ablation treatment session was modeled as either an outpatient or inpatient procedure, allowing for multiple RF ablation procedures over a lifetime as necessary. The direct costs of an outpatient procedure included reimbursements for hospital fees incurred by the RF ablation treatment session and observation room time, professional fees, laboratory fees, medication, and repeat abdominal CT at 1 month after the procedure. The components of cost associated with an inpatient RF ablation procedure differed only in that hospital reimbursement involved a global fee as determined by the Diagnosis Related Group assigned to the admission. This global fee incorporates hospital reimbursement for nonprofessional costs incurred during the inpatient admission, including laboratory, radiology, pharmacy, and procedural support. Costs of anesthesia and postprocedural medications for inpatient and outpatient procedures were incorporated into the model after weighting each cost by the frequency of need. Costs associated with the complications of RF ablation were based on the additional hospital and professional fees created by a complication of the procedure. Complication types and probability were derived from a large series by Livraghi et al (15) that summarized complications in 1,766 patients treated by percutaneous RF ablation, which afforded inclusion of both common and rare events. Complication costs were weighted by probability per treatment session and included the increased costs necessitated by an inpatient admission (for a planned outpatient procedure) or movement into a Diagnosis Related Group with a higher reimbursement rate (for a planned inpatient procedure). Other major costs associated with each complication, including professional fees and outpatient medication, were included as appropriate. The direct cost of regularly scheduled follow-up evaluations included reimbursements for the hospital and professional fees associated with an outpatient abdominal CT with and without intravenous contrast material,

Shetty et al

an outpatient office visit, and determination of a serum tumor marker. Importantly, the number of follow-up evaluations in our model was dependent on the overall median survival of patients treated with RF ablation (equal in our model to the median survival of palliative care plus marginal median survival of RF ablation). Survival in our population was modeled as a declining exponential approximation of life expectancy (16,17), and the number of required follow-up evaluations was set as the sum of the number of patients alive at each regularly programmed interval in the future. The direct costs of follow-up evaluations were discounted at a rate of 5% (18). Total direct costs of palliative care included the costs of the initial evaluation already described (abdominal CT with and without intravenous contrast material, an outpatient office visit, and determination of serum tumor marker) in addition to the costs of disease-related symptom control. Our model was simplified by the exclusion of indirect costs, including costs to the patient and to society, and exclusion of additional medical costs that might be incurred over a period of increased survival as a result of new or coexisting diseases. Calculation of Cost: Data Sources The model of direct costs associated with percutaneous RF ablation and palliative care required inputs associated with cost, probabilities of various outcomes, and survival. Table 2 details the components of cost associated with each treatment strategy as outlined earlier. Direct costs to the payer (Medicare) were derived from published 2000 Medicare reimbursement schedules for hospital and professional fees (19,20) appropriate to our institution and geographic location. These reimbursement rates incorporated components of the recently implemented Outpatient Prospective Payment System (OPPS) as appropriate. The costs to the payer associated with medications were based on the 2000 Health Care Financing Administration Upper Limit price (20) or on the average wholesale price (21). Baseline assumptions incorporated into our model were based on our internal patient series and the published literature. The independent variables included the baseline survival of pal-

•

825

liative care, marginal median survival of patients undergoing RF ablation, mean number of lifetime treatments per patient, frequency of follow-up evaluations, number of hours of observation, cost of each abdominal CT procedure with and without intravenous contrast material, probability of inpatient admission, probability of general anesthesia, probability of postprocedural medications, and discount rate. The baseline assumption for each of these variables is presented in Table 1. Calculation of Cost-Effectiveness The marginal cost-effectiveness of RF ablation was calculated as the marginal direct cost of RF ablation divided by the marginal median survival benefit conferred by the RF ablation treatment strategy. The model was constructed to allow calculation of cost-effectiveness over a range of marginal median survival benefits, from 0 to 60 months. In addition, a break-even analysis was performed to determine the required marginal survival benefit that would be required of the RF ablation treatment strategy to match benchmarks for costeffectiveness obtained from the literature: generous ($100,000/life year [LY] gained), moderate ($50,000/LY gained), and strict ($20,000/LY gained) (22,23). Sensitivity Analysis Sensitivity analysis was performed to test the influence of various parameters on the final cost-effectiveness calculation. The input parameters that were tested through this analysis included the number of lifetime treatments, number of hours of observation per treatment, frequency of follow-up evaluations, cost of each abdominal CT procedure, probability of general anesthesia, probability of postprocedural medication, discount rate, and method of imaging guidance. The model was also tested with two alternate scenarios of reimbursement: one based more closely on the new OPPS and a second that required inpatient admission for each RF ablation procedure.

RESULTS Literature Review Literature review revealed no published studies of the cost-effectiveness

826

•

Cost-effectiveness of RF Ablation for Hepatic Neoplasms

July 2001

JVIR

neous RF ablation at our institution between October 1998 and August 2000. Of these patients, 22 (47.8%) had HCC, 12 (26.1%) had CLM, and 12 (26.1%) had metastatic lesions from other primary sites. A total of 73 lesions were treated in 83 sessions for an average of 1.59 treatment sessions per patient. Records were reviewed for the use of general anesthesia (n ⫽ 3; 4.1%), the need for extended overnight observation as an outpatient (n ⫽ 7; 9.6%), and the need for inpatient admission (n ⫽ 0). Overall, the mean number of hours of observation required in our series was 4 hours, including those patients requiring extended overnight observation. Postprocedural medications were necessary in approximately 10% of RF ablation treatment sessions. Figure 1. Marginal cost of percutaneous RF ablation and cost-effectiveness of percutaneous RF ablation compared to palliative care, each varied over a range of marginal median survival of RF ablation between 1 and 60 months.

of RF ablation in the treatment of malignant hepatic neoplasms. The natural history of untreated HCC is highly dependent on the tumor burden (number and size of lesions), degree of liver impairment, and growth patterns of tumor (24). Cause of death in these patients is typically related to sequelae of underlying liver dysfunction, including liver failure and gastrointestinal bleeding (25). These factors create a heterogeneous group of patients when considering survival data, especially when considering that untreated patients tend to represent a selected group with a higher severity of neoplasms who were not suited for available treatments. Median survival of untreated HCC has been reported to be 2–4 months (26), although more recent series of patients who were unsuitable for resection or transplant describe 1-year survival rates of 54%–72% (27,28). In each study, there were clear effects on survival related to baseline tumor burden and liver functional status. This variability in observed survival is also likely influenced by lead time bias incurred by improving screening and detection techniques. Improved survival has been observed in patients treated with either surgical resection of tumor (5-year survival rates between 35% and 50% [29]) or orthotopic liver transplant (5-year survival rates between 36% and 81% [29]). However, patients treated with resection or transplant likely repre-

sent a highly selected subgroup with medical conditions and tumor burdens favorable to more aggressive treatment. The natural history of colorectal metastases is similarly difficult to examine. Independent determinants of survival in untreated patients include volume of tumor involvement, grade of malignancy, presence of extrahepatic disease, mesenteric lymph node involvement, serum carcinoembryonic antigen level, and age (10). Although surgical resection is considered the ideal treatment, exclusion criteria for surgical resection include bilobar hepatic involvement, more than four metastases, or tumor proximity to major vascular or biliary structures. Studies of median survival without treatment vary between 3.8 and 21.3 months (10,30 –32). The highest survival rates are seen in patients undergoing surgical resection, with 5-year survival rates between 25% and 39% (32); however, only a small subgroup of patients with CLM is suitable for surgery. Overall, untreated liver metastases are associated with an overall median survival of approximately 10 months (10,33). Patient Series Our patient series included 46 consecutive patients treated with percuta-

Calculation of Direct Costs As shown in Figure 1, the total marginal direct cost associated with the RF ablation treatment strategy over palliative care ranged between $9,158 and $17,461 when marginal median survival varied between 1 and 60 months. Marginal direct cost increased with additional survival because of an increasing number of requisite follow-up evaluations. The direct cost of each outpatient RF ablation treatment session was $4,321.98, which incorporates the direct costs of the procedure and weighted costs of complications, anesthesia, and postprocedural medications. The direct cost of each inpatient RF ablation treatment session was $7,181.67, which also included additional direct costs of complications, anesthesia, and postprocedural medications. Calculation of Cost-Effectiveness The cost-effectiveness of RF ablation compared to palliative treatment follows a hyperbolic curve because marginal median survival of RF ablation is varied between 1 and 60 months (Fig 1). The cost-effectiveness of the RF ablation treatment strategy at 6 months, 1 year, 2 years, 3 years, and 5 years marginal median survival benefit is $20,424, $11,407, $6,731, $5,034, and $3,492 per LY gained, respectively. A break-even analysis was performed to determine the marginal median survival benefit that must be generated by the RF ablation treatment strategy to achieve cost-effectiveness benchmarks ob-

Volume 12

Number 7

tained from the literature, compared to palliative care (22,23). This analysis revealed that to achieve strict ($20,000/LY gained), moderate ($50,000/LY gained), and generous ($100,000/LY gained) costeffectiveness thresholds, RF ablation would be required to generate 6.14 months, 2.26 months, and 1.10 months, respectively, of marginal median survival benefit compared to a palliative treatment strategy. These values are shown graphically in Figure 1. Sensitivity Analysis Sensitivity analysis was performed to determine the dependence of our model on the variables listed in Table 2. This analysis revealed strong dependence of cost-effectiveness on the mean number of lifetime RF ablation procedures (Fig 2), the number of hours of observation time per RF ablation treatment session (Fig 3), the frequency of follow-up evaluations, and the reimbursement rate for technical or professional fees for abdominal CT. As a result of recent changes in the reimbursement for outpatient procedures, we substituted into our model an alternate reimbursement scenario more consistent with the newly implemented Medicare OPPS. The OPPS involves payment of a global hospital fee for outpatient procedures, which includes hospital costs of the procedure, anesthesia, medications, and observation time (without including laboratory, radiology, or professional fees). This global fee is based on the assignment of each procedure to an Ambulatory Procedure Code (APC), each of which has a fixed rate of reimbursement (19). The historically accepted procedure code for RF ablation (Current Procedural Terminology code 49201) is now designated as an inpatient-only code under the OPPS, so it was necessary to assign an appropriate alternate APC, APC 0153 (Peritoneal and Abdominal Procedures). Under the OPPS, this APC is reimbursed at a rate $951.32 (19), and we included a technology pass-through charge for the disposable RF ablation electrodes ($595) consistent with Medicare policy. This substitution yielded a total direct cost of $3,482.79 per RF ablation treatment session. Results of analysis with this alternate scenario of outpatient RF ablation reim-

Shetty et al

bursement are shown in Table 1. Sensitivity analysis of this alternative reimbursement scheme was similar to the traditional reimbursement scheme, except RF ablation cost-effectiveness was no longer dependent on observation time, as this cost was now bundled into the global APC reimbursement rate. We also considered the effect of admitting all patients who undergo RF ablation to inpatient status. Under this scheme, cost per RF ablation treatment session increased to $7,181.67. Results of analysis with use of an inpatient RF ablation scenario are presented in Table 1.

DISCUSSION Our goal in this study was to evaluate the cost-effectiveness of percutaneous RF ablation over palliative care in the treatment of HCC and CLM. Although RF ablation has been successfully applied in other metastatic lesions, we decided to limit our analysis to these two forms of hepatic malignancy, for which there was relative availability of a large foundation of published survival and treatment data. Our analysis revealed that a percutaneous RF ablation treatment strategy would be required to produce only 6.14 months, 2.26 months, or 1.10 months, respectively, of marginal median survival to achieve strict ($20,000/LY gained), moderate ($50,000/LY gained), or generous ($100,000/LY gained) cost-effectiveness thresholds over palliative care. Cost-Effectiveness: Effect of Survival The overall survival of patients in the RF ablation treatment strategy arm of the model was composed of the baseline survival of palliative care plus the marginal survival of patients undergoing RF ablation. We elected to use a baseline median survival of palliative care of 10 months (10,27,28,33) for patients with HCC or CLM. This baseline assumption may represent a slight overestimation, particularly in the case of patients with HCC, resulting in a slight increase in the cost per LY gained. Determination of the exact survival benefit of RF ablation was made difficult by heterogeneity of lesion size, length of follow-up, and RF ablation technique reported in the literature (1– 4). In an early series by Rossi et al (1),

•

827

overall median survival of 44 months was reported for treatment of HCC. However, this series was limited to patients with small (⬍3 cm) lesions and employed conventional monopolar RF electrodes. Survival in RF-ablation–treated patients with CLM has been reported as 94%–100% at 1 year (6,7), 89% at 18 months (7), and 61.5% at 2 years (6), although these studies employed different RF ablation techniques and treated patients with different lesion sizes and tumor burdens (1,2,6 – 8). Rather than arbitrarily selecting a value for marginal median survival benefit accrued by RF ablation in patients with HCC or CLM, we adopted a more flexible strategy of the use of a range of marginal median survival between 1 and 60 months. Our analysis reveals that the costeffectiveness of RF ablation compared to palliative care is highly dependent on the marginal median survival of patients with RF ablation (Fig 1). Although the marginal cost of the RF ablation treatment strategy increases over time because of additional costs of follow-up, the strategy as a whole becomes relatively more cost-effective with increased median survival when considered as an expenditure per year of life gained by treatment. Cost-Effectiveness: Number of Lifetime Treatments The cost-effectiveness of RF ablation compared to palliative care was highly dependent on the number of lifetime treatments per patient (Table 3 and Fig 2). Increasing the number of lifetime treatments per patient reduced the costeffectiveness of RF ablation compared to palliative care. Specifically, RF ablation would be required to create a marginal median survival benefit of 4.40, 7.36, 10.30, 13.24, and 16.14 months to achieve a strict ($20,000/LY gained) cost-effectiveness threshold with 1, 2, 3, 4, and 5 lifetime RF ablation treatment sessions. The type of malignant liver neoplasm will likely influence the requisite number of lifetime RF ablation treatments. For example, the presence of a fibrous tumor capsule in HCC protects adjacent liver tissue from damage during RF ablation. Although this will increase success with a particular HCC lesion, the disease typically occurs in cirrhotic liver, creating

828

•

Cost-effectiveness of RF Ablation for Hepatic Neoplasms

July 2001

JVIR

Table 2 Determinants of Direct Cost for Percutaneous RF Ablation and Palliative Care Treatment Strategies* Direct Cost Initial Evaluation Physician office visit Physician office visit professional fee Abdominal CT examination Serum tumor marker measurement Carcinoembryonic Antigen Alpha Fetoprotein Outpatient RFA procedure Hospital reimbursement Observation time (per hour) Abdominal CT examination US guidance Professional fee: destruction of intraabdominal lesion Professional fee: CT abdomen with and without intravenous contrast Professional fee: US guidance Kefzol 1 g intravenous Preprocedural laboratory Prothrombin time Partial thromboplastin time Automated complete blood count Creatinine Postprocedural imaging Abdominal CT examination Professional fee: abdominal CT with and without intravenous contrast Inpatient RF ablation procedure: 0% of treatment sessions‡ Hospital reimbursement Professional fee: destruction of intraabdominal lesion Professional fee: CT abdomen with and without intravenous contrast Professional fee: US guidance Kefzol 1 g intravenous Postprocedural imaging Abdominal CT examination Professional fee: abdominal CT with and without intravenous contrast General anesthesia: 4.1% of treatment sessions‡ Professional charges Lidocaine 50 cm3 Postprocedural medications: 10% of treatment sessions‡ 1 tablet Acetaminophen 325 mg/Oxycodone 5 mg Complications (rate per patient)㛳 Liver failure (0.170%) Hemothorax (0.057%)

Value ($) 80.49 135.10 237.10 26.22 23.18 1413.09 244.00 237.10 108.13 1066.07 78.03 38.09 2.54 5.43 8.30 7.72 7.07 237.10 78.03

5656.24 1066.07 78.03

APC 602 (High Level Clinic Visit) CPT 99205 APC 283 (Level II Computerized Axial Tomography) CPT 82378 CPT 82105 CPT 49201 Revenue Code 762† APC 283 APC 268 (Guidance under Ultrasound) CPT 49201 CPT 74170–26 CPT 76429–26 CPT J0690 CPT CPT CPT CPT

85610 85730 85021 82565

APC 283 CPT 74170–26

DRG 203 (Malignancy of hepatobiliary system or pancreas)§ CPT 49201 CPT 74170–26

38.09 2.54

CPT 76429–26 CPT J0690

237.10 78.03

APC 283 CPT 74170–26

260.54 0.95

CPT 00702 (4 Base Units ⫹ 10 Time Units) CPT J2000

0.82 5717.80 5911.02

Hepatic abscess (0.057%) Pulmonary embolism (0.057%)

6014.16 6065.67

Biloma formation (0.057%) Acute cholecystitis (0.057%) Septic shock (0.057%) Hemorrhage requiring surgery (0.057%)

6439.76 6462.86 6498.21 8844.96

Intestinal perforation (0.170%)

9202.22

Skin burn at grounding pad site (0.283%)¶ Subcutaneous/peritoneal seeding (0.227%)# Follow-up evaluation** Physician office visit Physician office visit professional fee Abdominal CT examination

2000 Medicare Reimbursement Code (19,20)

5.55 80.49 91.93 237.10

(21) (Assumes 1–2 tablet every 4–6 hours for 4 days) DRG 202 (Cirrhosis and Alcoholic Hepatitis) DRG 203 (Malignancy of Hepatobiliary System or Pancreas) and CPT 32020 DRG 203, CPT 47011 & CPT 76934–26 DRG 078 (Pulmonary Embolism) and CPT 78588–26 (Ventilation/Perfusion Scan, Professional Fee) DRG 203, CPT 47511 & CPT 75980–26 DRG 203, CPT 47563 (Laparoscopic Cholecystectomy) DRG 416 (Septicemia, age ⬎17) DRG 192 (Pancreas, Liver, Shunt Procedures) and CPT 47350 (Simple Liver Repair) DRG 152 (Minor Small & Large Bowel Procedures with complication or comorbidity) and CPT 49002 (Laparotomy) Silver Sulfadiazene 1% Cream, 50 g APC 602 CPT 99215 APC 283

Volume 12

Number 7

Shetty et al

•

829

Table 2 (Continued) Direct Cost Serum tumor marker measurement Carcinoembryonic antigen Alpha fetoprotein Palliative Treatment Initial Evaluation Physician office visit Physician office visit professional fee Abdominal CT examination Serum tumor marker measurement Carcinoembryonic Antigen Alpha Fetoprotein Symptom Control††

Value ($)

2000 Medicare Reimbursement Code (19,20)

26.22 23.18

CPT 82378 CPT 82105

80.49 135.10 237.10

APC 602 CPT 99205 APC 283

26.22 23.18 x

CPT 82378 CPT 82105

Note.—CPT ⫽ Current Procedure Terminology; DRG ⫽ Diagnosis Related Group. * Median survival of palliative care: 10 months (see Discussion). Marginal median survival of RF ablation: 34 months (1). Mean number of treatments per patient: 1.59 (from internal patient series). These baseline values were used only if not explicitly varied in cost-effectiveness analysis. † Mean observation time per treatment: 4.0 hours (from internal patient series). ‡ Derived from internal patient series (see Results). § Inpatient Prospective Payment System reimbursement based on DRG blended rate of $4,335.61 㛳 Rates of complication from (15). Rates noted are per patient. Costs noted are substituted for the hospital reimbursement noted in the inpatient RFA treatment. All complications were expected to result in inpatient admission except as noted. Complications (with rates) not expected to alter hospital reimbursement in DRG 203: Hemorrhage requiring transfusion (0.283%), Hepatic requiring no intervention (0.057%), Diaphragmatic injury (0.057%), Bradycardia (0.057%), Multisegmental hepatic infarction (0.057%). ¶ Medication cost added to cost of procedure. No inpatient admission required unless already planned. # No immediate cost of complication. No inpatient admission required unless already planned. ** Frequency of evaluations: 3 months (from internal patient series). Discount rate for cost of follow-up evaluations: 5% (18). †† Disease-related symptom control valued at dummy variable x because cost similar between RFA and palliative care treatment strategies; there is therefore no effect on marginal cost (22).

a substrate prone to development of metachronous lesions that will require additional RF ablation treatment sessions (5,34). In contrast, CLM tumors are hard and tend to have irregular borders, requiring additional ablation of a 0.5–1-cm tumor-free margin around the lesion. Therefore, patients with CLM treated with RF have had a high recurrence rate, which has reduced efficacy and necessitated additional treatments of each lesion (5,34). The number of lifetime RF ablation treatments may also decrease with evolution and improvement in RF ablation technology. Technologic innovations have increased the size of treatable lesions from 1.6 cm achieved with older monopolar simple needles (34) to larger volumes of coagulation necrosis currently achieved in vivo with the use of internally-cooled electrodes (7,8), multiple array hook electrodes (1), multiprobe arrays (8), and minimization of perfusion-mediated heat dissipation (35,36). Whereas an early series (1) required 3.3 treatments per patient, our series used a mean of

only 1.59 treatments per patient. However, given the nature of these diseases, we anticipate that this initial reduction in the number of treatments may be balanced by increased survival and future application of RF ablation to previously undetected or metachronous hepatic lesions. Sensitivity Analysis Our model incorporates assumptions based on the specific application of RF ablation at our institution, which is likely to differ from methods used elsewhere. Percutaneous RF ablation is performed entirely on an outpatient basis with infrequent (4.1%) use of general anesthesia, an average of 4 hours of observation time per patient, and outpatient follow-up every 3 months. Our sensitivity analysis revealed that use of general anesthesia had little effect on the cost-effectiveness calculation (Table 3), whereas the performance of RF ablation as an inpatient (Table 1) would be expected to reduce RF ablation cost-effectiveness

compared to palliative care. Our baseline assumption of 4 hours of observation time is derived from our patient experience and has been reported elsewhere (37). Our analysis revealed that the cost-effectiveness calculation was highly dependent on observation time (Table 3 and Fig 3). Our choice of frequency of follow-up (every 3 months) is based on our own clinical practice and has also been reported elsewhere (37). The cost-effectiveness of RF ablation compared to palliative care was highly dependent on follow-up interval (Table 3), although there is little evidence that more frequent follow-up evaluation would provide additional benefit. Selection of imaging modality can play an important role in the cost of the RF ablation treatment strategy. Guidance of therapy can be completed with US, CT, or MR imaging, either alone or in combination. We employed CT fluoroscopy and US guidance to direct the RF ablation probe and monitor therapy during the procedure, and employed contrast-enhanced abdomi-

830

•

Cost-effectiveness of RF Ablation for Hepatic Neoplasms

July 2001

JVIR

Our analysis reveals that the cost-effectiveness of RF ablation compared to palliative care was sensitive to the reimbursement for technical and professional fees of abdominal CT, likely because the examination is an integral part of both the RF ablation procedure and the follow-up evaluations. It may be possible to further reduce the direct cost of each RF ablation treatment by employing an US microbubble contrast agent (40) to screen patients for residual tumor. Reimbursement for RF Ablation

Figure 2. Cost-effectiveness of percutaneous RF ablation compared to palliative care varied over a range of marginal median survival of RF ablation between 1 and 60 months. Each curve represents the number of lifetime treatments per patient as part of the RF ablation treatment strategy and focuses on cost-effectiveness values below $30,000/LY gained.

Figure 3. Cost-effectiveness of percutaneous RF ablation compared to palliative care varied over a range of mean number of hours of observation per treatment session, between 0 and 24 hours. Each curve represents the number of lifetime treatments per patient as part of the RF ablation treatment strategy.

nal CT immediately after the procedure to gauge efficacy and allow for possible reintervention while the patient was still on the procedure table. Our sensitivity analysis (Table 3) shows that there was little effect on the overall cost of the RF ablation treat-

ment strategy when comparing the more expensive combination of CT and US guidance with the least expensive option, US guidance alone. CT and MR imaging are the preferred modalities for monitoring efficacy of RF ablation treatment (38,39).

Although we have used the historically accepted method of billing for percutaneous RF ablation, the everchanging landscape of reimbursement policy continues to burden the clinician and complicate cost analysis. Our model was based on a traditional mechanism of reimbursement that incorporated elements of the new OPPS system with elements of the older costbased reimbursement system. The previously accepted Current Procedural Terminology code for percutaneous RF ablation (49201) is now designated as an inpatient-only code and will not be reimbursed by Medicare if performed as an outpatient (19,20). In this environment, the provider is left with few options until the inpatient-only designation is removed or percutaneous RF ablation is specifically placed on the OPPS reimbursement schedule. In fact, the current system might create a perverse incentive to admit patients and perform RF ablation on an inpatient basis. However, our analysis shows that this would actually result in a more costly alternative for the payer and decrease the cost-effectiveness of RF ablation (Table 1). The cost of each RF ablation treatment would increase from $4,321.98 per treatment to $7,181.67 per treatment and increase the requisite marginal median survival benefit of RF ablation from 6.14 months to 9.25 months to meet the strict threshold of cost-effectiveness ($20,000/LY gained). A hypothetical alternate reimbursement scheme based more closely on the OPPS resulted in a decrease in reimbursement to $3,482.79 per treatment session and yielded a decreased marginal median survival benefit of RF ablation of 5.23 months required to achieve a strict threshold of cost-effectiveness from

Volume 12

Number 7

Shetty et al

•

831

Table 3 Sensitivity Analyses

Range of Values

Variable Baseline Mean number of treatments per patient (lifetime)

1.0–5.0

Observation time per treatment (h)

0–24

Frequency of evaluations (mo)

1–6

Technical charge, outpatient abdominal CT Professional fee, abdominal CT with and without intravenous contrast material Probability of general anesthesia (per treatment)

$100–$400 $30–$100 0.0–1.0

Probability of postprocedural medications (per treatment)

0.0–1.0

Discount rate

0%–10%

RF ablation treatment guidance (% of sessions) 100% both CT and US guidance 50% CT and US, 50% US only 100% US guidance only

Break-Even Analysis†

Value

Marginal Direct Cost of RFA versus Palliative Care ($)*

$100,000/ LY Gained

$50,000/ LY Gained

$20,000/ LY Gained

1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 0 8 24 1 6 $100 $400 $30 $100 0.0 0.5 1.0 0.0 0.5 1.0 0% 10%

9,158–17,461 6,608–14,911 8,769–17,072 10,930–19,233 13,091–21,394 15,252–23,555 17,413–25,716 19,574–27,877 21,735–30,038 23,896–32,199 7,624–15,927 10,693–18,995 16,830–25,132 14,035–25,482 7,900–12,667 8,115–14,202 10,398–21,333 8,792–16,318 9,326–17,984 9,141–17,444 9,348–17,651 9,556–17,858 9,152–17,455 9,183–17,486 9,215–17,517 9,314–20,166 9,025–15,605

1.10 0.79 1.05 1.32 1.59 1.85 2.12 2.38 2.65 2.92 0.91 1.29 2.05 1.73 0.95 0.97 1.26 1.06 1.12 1.10 1.12 1.15 1.10 1.10 1.11 1.12 1.08

2.26 1.62 2.16 2.71 3.26 3.80 4.35 4.90 5.44 5.99 1.88 2.65 4.20 3.71 1.92 1.98 2.60 2.16 2.31 2.26 2.31 2.36 2.26 2.27 2.28 2.31 2.22

6.14 4.40 5.88 7.36 8.83 10.30 11.77 13.24 14.69 16.14 5.09 7.19 11.37 11.07 4.99 5.27 7.28 5.83 6.29 6.13 6.28 6.42 6.14 6.16 6.18 6.38 5.97

9,158–17,461 8,911–17,213 8,663–16,966

1.10 1.07 1.04

2.26 2.20 2.14

6.14 5.98 5.81

* Range: 1– 60 mo RFA marginal median survival. † Months marginal median survival with RFA compared to palliative care.

the perspective of the payer (Table 1). A related issue is the dependence of the cost-effectiveness of RF ablation on observation time, which becomes more pronounced as the number of RF ablation treatments increases (Fig 3). Important, this component of variation drops out of the calculation if the alternate scenario of outpatient reimbursement (based on the OPPS) is employed, because observation time is considered part of the global reimbursement determined by APC group. Although this may reduce costs from the perspective of the payer, the provider will receive a fixed reimbursement regardless of changing requirements for observation time. Because the cost-effectiveness of RF ablation compared to palliative care is highly

dependent on observation time in the traditional reimbursement scenario, it will be crucial that providers continue to report accurate observation room times to ensure that a future OPPS global fee for RF ablation accurately compensates for the provider’s true costs. These issues highlight the importance of ensuring that reimbursement policies are in accordance with the rapidly evolving realities of percutaneous RF ablation technology. Several aspects of our model deserve further consideration. First, our analysis depends on direct costs to the payer (Medicare) and does not include indirect costs of the RF ablation treatment strategy, including costs to society. Although our decision to employ Medicare reimbursement data as

a substitute for more detailed cost accounting methods is an accepted alternative in determining direct costs (22,41), this method does not consider the perspective of the provider and the differences between actual costs to the provider and reimbursement rates. Finally, although our model does not adjust for quality of life, our experience indicates that the RF ablation procedure creates minimal disruption for the patient and is associated with an extremely short recovery period. Our work focused only on percutaneous RF ablation as a treatment modality for HCC and CLM. Although this limitation was vital to the creation of a focused model to assess the costeffectiveness of percutaneous RF ablation compared to palliative care, part

832

•

Cost-effectiveness of RF Ablation for Hepatic Neoplasms

of the promise of the technique lies in its use as part of an armamentarium for patients with both HCC and CLM, possibly in conjunction with other modalities (11,42). Percutaneous ethanol injection has been shown to be equal to surgical resection in terms of survival results (43), and RF ablation has been shown to be more effective than percutaneous ethanol injection in the treatment of individual lesions (3). Percutaneous RF ablation may therefore hold promise as part of a combination therapy or even as first-line therapy in patients for whom hepatic functional reserve is limited, as in the case of cirrhosis. We currently continue to recommend surgical treatment when possible, although a future determination of the long-term equivalence of RF ablation and surgery would allow direct cost-effectiveness comparison between the techniques. In conclusion, we have sought to evaluate the cost-effectiveness of percutaneous RF ablation when compared to a palliative approach in the treatment of HCC and CLM. With use of a range of possible survival outcomes for patients undergoing the treatment, RF ablation would be required to generate a marginal median survival benefit of 6.14 months, 2.26 months, and 1.10 months to achieve strict ($20,000/LY gained), moderate ($50,000/LY gained), and generous ($10,0000/LY gained) cost-effectiveness thresholds, respectively. These survival results are modest and have likely already been surpassed by the current use of RF ablation, making the treatment an attractive strategy both medically and economically. This cost-effectiveness determination was sensitive to the marginal median survival created by RF ablation and the required number of lifetime RF ablation treatment sessions (Fig 2), two clinical variables for which values depend on the future performance of prospective, randomized trials of percutaneous RF ablation. Despite the current lack of long-term survival and treatment data, our analysis permits a flexible and accurate approach to the use of cost-effectiveness information for this new technology and allows application of this data to situations of clinical uncertainty. Indeed, our review of the literature shows heterogeneity in the types of patients being treated with percutaneous RF

ablation. As survival and treatment data from particular subgroups of patients are generated, it will be possible to apply this analysis to allow accurate determination of cost-effectiveness in smaller, more homogeneous populations of patients with HCC or CLM. Our study also reveals the importance of reimbursement mechanism and that reimbursement policy keeps pace with RF ablation technology. References 1. Rossi S, Di Stasi M, Buscarini E, et al. Percutaneous RF interstitial thermal ablation in the treatment of hepatic cancer. AJR Am J Roentgenol 1996; 167: 759 –768. 2. Rossi S, Buscarini E, Garbagnati F, et al. Percutaneous treatment of small hepatic tumors by an expandable RF needle electrode. AJR Am J Roentgenol 1998; 170:1015–1022. 3. Livraghi T, Goldberg SN, Lazzaroni S, Meloni F, Solbiati L, Gazelle GS. Small hepatocellular carcinoma: treatment with radio-frequency ablation versus ethanol injection. Radiology 1999; 210:655– 661. 4. Livraghi T, Goldberg SN, Lazzaroni S, et al. Hepatocellular carcinoma: radio-frequency ablation of medium and large lesions. Radiology 2000; 214:761– 768. 5. Jiao LR, Hansen PD, Havlik R, Mitry RR, Pignatelli M, Habib N. Clinical short-term results of radiofrequency ablation in primary and secondary liver tumors. Am J Surg 1999; 177:303– 306. 6. Solbiati L, Ierace T, Goldberg SN, et al. Percutaneous US-guided radio-frequency tissue ablation of liver metastases: treatment and follow-up in 16 patients. Radiology 1997; 202:195–203. 7. Solbiati L, Goldberg SN, Ierace T, et al. Hepatic metastases: percutaneous radio-frequency ablation with cooled-tip electrodes. Radiology 1997; 205:367– 373. 8. Goldberg SN, Solbiati L, Hahn PF, et al. Large-volume tissue ablation with radio frequency by using a clustered, internally cooled electrode technique: laboratory and clinical experience in liver metastases. Radiology 1998; 209: 371–379. 9. Siperstein AE, Rogers SJ, Hansen PD, Gitomirsky A. Laparoscopic thermal ablation of hepatic neuroendocrine tumor metastases. Surgery 1997; 122:1147–1154; discussion, 1154 –1155. 10. Stangl R, Altendorf-Hofmann A, Charnley RM, Scheele J. Factors influencing the natural history of colorectal liver metastases. Lancet 1994; 343:1405–1410.

July 2001

JVIR

11. Rose DM, Allegra DP, Bostick PJ, Foshag LJ, Bilchik AJ. Radiofrequency ablation: a novel primary and adjunctive ablative technique for hepatic malignancies. Am Surg 1999; 65:1009 – 1014. 12. Goldberg SN, Gazelle GS, Mueller PR. Thermal ablation therapy for focal malignancy: a unified approach to underlying principles, techniques, and diagnostic imaging guidance. AJR Am J Roengenol 2000; 174:323–331. 13. Goldberg SN, Gazelle GS, Compton CC, Mueller PR, Tanabe KK. Treatment of intrahepatic malignancy with radiofrequency ablation: radiologicpathologic correlation. Cancer 2000; 88: 2452–2463. 14. Goldberg SN, Stein MC, Gazelle GS, Sheiman RG, Kruskal JB, Clouse ME. Percutaneous radiofrequency tissue ablation: optimization of pulsed- radiofrequency technique to increase coagulation necrosis. J Vasc Interv Radiol 1999; 10:907–916. 15. Livraghi T, Solbiati L, Meloni F, Ierace T, Goldberg SN, for the Italian Multicenter Cooled-tip RF Study Group. Complications after cool-tip RF ablation of liver cancer: initial report of the Italian Multicenter Cooled-tip RF Study Group [abstract]. Radiology 2000; 217(suppl):27. 16. Beck JR, Kassirer JP, Pauker SG. A convenient approximation of life expectancy (the “DEALE”). I. validation of the method. Am J Med 1982; 73:883– 888. 17. Beck JR, Pauker SG, Gottlieb JE, Klein K, Kassirer JP. A convenient approximation of life expectancy (the “DEALE”). II: use in medical decisionmaking. Am J Med 1982; 73:889 – 897. 18. Petitti DB. Meta-analysis, decision analysis, and cost-effectiveness analysis: methods for quantitative synthesis in medicine. Monographs in Epidemiology and Biostatistics. New York: Oxford University Press, 1994. 19. Health Care Financing Administration Web Site. Available at http://www. hcfa.gov/medicare/hopsmain.htm. Accessed September 2000. 20. Medicare B Bulletin. National Heritage Insurance Company. Available at http://www.medicarenhic.com. Accessed September 2000. 21. 1999 Drug Topics Red Book. Montvale, NJ: Medical Economics, 1999. 22. Abramson RG, Rosen MP, Perry LJ, Brophy DP, Raeburn SL, Stuart KE. Cost-effectiveness of hepatic arterial chemoembolization for colorectal liver metastases refractory to systemic chemotherapy. Radiology 2000; 216: 485– 491. 23. Smith TJ, Hillner BE, Desch CE. Efficacy and cost-effectiveness of cancer

Volume 12

24.

25.

26.

27.

28.

29. 30.

Number 7

treatment: rational allocation of resources based on decision analysis. J Natl Cancer Inst 1993; 85:1460 –1474. Colombo M. The natural history of hepatocellular carcinoma in Western countries. Hepatogastroenterology 1998; 45(suppl 3):1221–1225. Okuda K, Ohtsuki T, Obata H, et al. Natural history of hepatocellular carcinoma and prognosis in relation to treatment: atudy of 850 patients. Cancer 1985; 56:918 –928. Stuart KE, Anand AJ, Jenkins RL. Hepatocellular carcinoma in the United States: prognostic features, treatment outcome, and survival. Cancer 1996; 77:2217–2222. Llovet JM, Bustamante J, Castells A, et al. Natural history of untreated nonsurgical hepatocellular carcinoma: rationale for the design and evaluation of therapeutic trials. Hepatology 1999; 29:62–67. Villa E, Moles A, Ferretti I, et al. Natural history of inoperable hepatocellular carcinoma: estrogen receptors’ status in the tumor is the strongest prognostic factor for survival. Hepatology 2000; 32:233–238. Nakakura EK, Choti MA. Management of hepatocellular carcinoma. Oncology (Huntingt) 2000; 14:1085–1098. Scheele J, Stangl R, Altendorf-Hofmann A. Hepatic metastases from

Shetty et al

31.

32.

33.

34. 35.

36.

37.

colorectal carcinoma: impact of surgical resection on the natural history. Br J Surg 1990; 77:1241–1246. Luna-Perez P, Rodriguez-Coria DF, Arroyo B, Gonzalez-Macouzet J. The natural history of liver metastases from colorectal cancer. Arch Med Res 1998; 29:319 –324. Ballantyne GH, Quin J. Surgical treatment of liver metastases in patients with colorectal cancer. Cancer 1993; 71(suppl):4252– 4266. Lehnert T, Otto G, Herfarth C. Therapeutic modalities and prognostic factors for primary and secondary liver tumors. World J Surg 1995; 19:252–263. Solbiati L. New applications of ultrasonography: interventional ultrasound. Eur J Radiol 1998; 27(suppl 2):S200–S206. Goldberg SN, Gazelle GS, Dawson SL, Rittman WJ, Mueller PR, Rosenthal DI. Tissue ablation with radiofrequency using multiprobe arrays. Acad Radiol 1995; 2:670 – 674. Goldberg SN, Hahn PF, Halpern EF, Fogle RM, Gazelle GS. Radio-frequency tissue ablation: effect of pharmacologic modulation of blood flow on coagulation diameter. Radiology 1998; 209:761–767. Rhim H, Dodd GD III. Radiofrequency thermal ablation of liver tumors. J Clin Ultrasound 1999; 27:221–229.

•

833

38. Goldberg SN, Gazelle GS, Solbiati L, et al. Ablation of liver tumors using percutaneous RF therapy. AJR Am J Roentgenol 1998; 170:1023–1028. 39. Sironi S, Livraghi T, Meloni F, De Cobelli F, Ferrero C, Del Maschio A. Small hepatocellular carcinoma treated with percutaneous RF ablation: MR imaging follow-up. AJR Am J Roentgenol 1999; 173:1225–1229. 40. Solbiati L, Goldberg SN, Ierace T, Dellanoce M, Livraghi T, Gazelle GS. Radio-frequency ablation of hepatic metastases: postprocedural assessment with a US microbubble contrast agent— early experience. Radiology 1999; 211:643– 649. 41. Neuman ML, Murphy BD, Rosen MP. Bedside placement of peripherally inserted central catheters: a cost- effectiveness analysis. Radiology 1998; 206: 423– 428. 42. Kainuma O, Asano T, Aoyama H, et al. Combined therapy with radiofrequency thermal ablation and intra-arterial infusion chemotherapy for hepatic metastases from colorectal cancer. Hepatogastroenterology 1999; 46:1071–1077. 43. Livraghi T, Giorgio A, Marin G, et al. Hepatocellular carcinoma and cirrhosis in 746 patients: long-term results of percutaneous ethanol injection. Radiology 1995; 197:101–108.