Journal of Surgical Oncology 2014;109:516–520
Cost-Effectiveness of Simultaneous Resection and RFA Versus 2-stage Hepatectomy for Bilobar Colorectal Liver Metastases DANIEL E. ABBOTT, MD,1* VANCE Y. SOHN, MD,2 DENNIS HANSEMAN, PhD,3 AND STEVEN A. CURLEY, MD4 1 Department of Surgery, University of Cincinnati, Cincinnati, Ohio Department of Surgical Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas 3 Department of Surgery, University of Cincinnati, Cincinnati, Ohio 4 Department of Surgical Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas 2
Background and Objectives: The current healthcare climate demands evaluation of treatment modalities in terms of costs and benefits. We compared the cost‐effectiveness of two different strategies for bilobar colorectal liver metastases (bCRLM). Methods: Patients with bCRLM treated with either resection/RFA or planned 2‐stage hepatectomy at our institution between 1999 and 2011 were reviewed. A decision analysis model was populated with treatment probabilities, outcomes, survival, and costs (Medicare payment, 2011 US$). Results: Two hundred fourteen patients underwent resection/RFA. Eighty‐two patients were treated with planned 2‐stage hepatectomy; 26 (32%) patients never completed a 2nd resection. In the 2‐stage cohort, 50 patients underwent portal vein embolization (PVE). Overall complication rate and 90‐day mortality for resection/RFA was 36% and 3.7%, and for 2‐stage hepatectomy (both procedures combined) was 44% and 7.3%, respectively. Cost‐effectiveness analysis revealed that resection/RFA cost $37,120 for 46.2‐month survival, while planned 2‐stage resection cost $62,198 for 35.9‐ month survival. If, hypothetically, all 2‐stage patients completed both stages of resection, the per‐patient cost was $72,644 for 40.3‐month survival. Conclusions: Resection/RFA is associated with lower costs and longer survival when compared to 2‐stage resection. This 1‐stage approach for bCRLM should be viewed as an efficient use of resources for this challenging clinical scenario.
J. Surg. Oncol. 2014;109:516–520. ß 2013 Wiley Periodicals, Inc.
KEY WORDS: liver surgery; comparative effectiveness; resource utilization
INTRODUCTION Historically, bilobar liver metastases from colorectal cancer (bCRLM) were a contraindication to hepatic metastasectomy. Over the past 2 decades, however, the introduction and evolution of both ablative techniques (radiofrequency ablation (RFA), microwave energy and irreversible electroporation) and technologic advancements that enhance resection (CUSA, Habib, Harmonic scalpel, and portal vein embolization—PVE) have advanced liver surgery [1–5]. As a result of this sophistication, an increasing number of patients with bCRLM are being considered for liver directed therapy [4,6,7]. Depending on the size and location of metastatic lesions in either hemi‐liver, both ablative techniques and resection can be considered for definitive therapy of these tumors [1,4,8]. Advocates for a resection‐only strategy would highlight large retrospective case series that demonstrate favorable survival for patients undergoing hepatic resection as opposed to ablation for colorectal liver metastases (CRLM), believing that resection is always a better therapeutic approach than ablation [9–12]. Furthermore, there is growing evidence that 2‐stage hepatectomy is a safe and effective strategy for bCRLM [7,11]. Conversely, however, practitioners who subscribe to the efficacy of ablative techniques argue that no randomized studies exist to conclude that resection is better than ablation for CRLM, and that a combined resection/RFA approach to bCRLM results in less morbidity than two (or more) hepatic resections [1,13–15]. This clinical uncertainly is unfolding in an increasingly difficult healthcare climate. Many patients and practitioners continue to push for more health care at a time when healthcare costs, and specifically cancer care expenditures, are rising at an unsustainable rate [16,17]. These opposing forces require that clinicians, researchers and policy makers consider both cost and effectiveness in designing treatment strategies for patients [18,19]. Ultimately, for patients with bCRLM, the optimal treatment strategy has not been clearly defined. Randomized data do not exist to guide our
ß 2013 Wiley Periodicals, Inc.
clinical management, and for this reason cost must be included as a metric by which various treatment strategies are measured [19,20]. We hypothesized that because surgical intervention is a costly endeavor, 1‐ stage hepatic resection with RFA was a more cost‐effective strategy than 2‐stage hepatectomy for patients with bCRLM. Furthermore, we sought to determine if there were any peri‐operative or oncologic outcome differences.
MATERIALS AND METHODS Patient Selection The University of Texas MD Anderson Cancer Center Institutional Review Board approved a retrospective review of all patients with bCRLM who were treated with either 1‐stage (resection/RFA) or 2‐stage hepatectomy. Inclusion criteria were patients over 18 years of age with either metachronous or synchronous bCRLM. Exclusion criteria included patients who underwent any resection or ablation at an outside institution or who had their bCRLM resected in one setting. Patient‐specific variables including gender, location of primary tumor, metachronous versus synchronous disease, BMI, use of chemotherapy, size of hepatic lesions, extent of resection, resection margin, the presence of absence of PVE, repeat PVE, post‐operative morbidity/mortality,
The authors have no financial support to report. *Correspondence to: Daniel Abbott, MD, 234 Goodman St, ML 0772, Cincinnati, OH 45219. Fax: 513‐584‐0459. E‐mail: [email protected] Received 17 August 2013; Accepted 30 November 2013 DOI 10.1002/jso.23539 Published online 23 December 2013 in Wiley Online Library (wileyonlinelibrary.com).
Cost‐Effectiveness Liver Metastases local and/or distant recurrence, recurrence‐free survival and overall survival were collected and analyzed. RFA was performed according to previously published reports [10,21]. Briefly, all patients were treated during open laparotomy using ultrasound guided RF needle placement and lesions were ablated with the RF 2000 or RF 3000 generator system (Boston Scientific, Natick, MA). The treatment goal was always a 1 cm zone of normal parenchymal destruction around the tumor. Depending on lesion size, initial power was 50 W (3.5 cm array) or 80 W (4 cm array) and increased in 10 W increments at 1‐min intervals. For lesions >2.5 cm, the array was positioned multiple times.
Statistical Analysis Statistical analysis was performed with SAS Version 9.3 (Cary, NC). Continuous variables were compared using the Wilcoxon rank sum test. Categorical variables were compared using standard chi‐square analysis or Fisher’s exact test, as appropriate. Overall survival was calculated using the method of Kaplan and Meier, from the date of surgery (first surgery if 2‐stage approach) to the date of death. Recurrence‐free survival was calculated from the date of surgery to the date of cancer recurrence, either locoregional or systemic, or the date of death from another cause. A P‐value of
Cost-Effectiveness of Simultaneous Resection and RFA Versus 2-stage Hepatectomy for Bilobar Colorectal Liver Metastases DANIEL E. ABBOTT, MD,1* VANCE Y. SOHN, MD,2 DENNIS HANSEMAN, PhD,3 AND STEVEN A. CURLEY, MD4 1 Department of Surgery, University of Cincinnati, Cincinnati, Ohio Department of Surgical Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas 3 Department of Surgery, University of Cincinnati, Cincinnati, Ohio 4 Department of Surgical Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas 2
Background and Objectives: The current healthcare climate demands evaluation of treatment modalities in terms of costs and benefits. We compared the cost‐effectiveness of two different strategies for bilobar colorectal liver metastases (bCRLM). Methods: Patients with bCRLM treated with either resection/RFA or planned 2‐stage hepatectomy at our institution between 1999 and 2011 were reviewed. A decision analysis model was populated with treatment probabilities, outcomes, survival, and costs (Medicare payment, 2011 US$). Results: Two hundred fourteen patients underwent resection/RFA. Eighty‐two patients were treated with planned 2‐stage hepatectomy; 26 (32%) patients never completed a 2nd resection. In the 2‐stage cohort, 50 patients underwent portal vein embolization (PVE). Overall complication rate and 90‐day mortality for resection/RFA was 36% and 3.7%, and for 2‐stage hepatectomy (both procedures combined) was 44% and 7.3%, respectively. Cost‐effectiveness analysis revealed that resection/RFA cost $37,120 for 46.2‐month survival, while planned 2‐stage resection cost $62,198 for 35.9‐ month survival. If, hypothetically, all 2‐stage patients completed both stages of resection, the per‐patient cost was $72,644 for 40.3‐month survival. Conclusions: Resection/RFA is associated with lower costs and longer survival when compared to 2‐stage resection. This 1‐stage approach for bCRLM should be viewed as an efficient use of resources for this challenging clinical scenario.
J. Surg. Oncol. 2014;109:516–520. ß 2013 Wiley Periodicals, Inc.
KEY WORDS: liver surgery; comparative effectiveness; resource utilization
INTRODUCTION Historically, bilobar liver metastases from colorectal cancer (bCRLM) were a contraindication to hepatic metastasectomy. Over the past 2 decades, however, the introduction and evolution of both ablative techniques (radiofrequency ablation (RFA), microwave energy and irreversible electroporation) and technologic advancements that enhance resection (CUSA, Habib, Harmonic scalpel, and portal vein embolization—PVE) have advanced liver surgery [1–5]. As a result of this sophistication, an increasing number of patients with bCRLM are being considered for liver directed therapy [4,6,7]. Depending on the size and location of metastatic lesions in either hemi‐liver, both ablative techniques and resection can be considered for definitive therapy of these tumors [1,4,8]. Advocates for a resection‐only strategy would highlight large retrospective case series that demonstrate favorable survival for patients undergoing hepatic resection as opposed to ablation for colorectal liver metastases (CRLM), believing that resection is always a better therapeutic approach than ablation [9–12]. Furthermore, there is growing evidence that 2‐stage hepatectomy is a safe and effective strategy for bCRLM [7,11]. Conversely, however, practitioners who subscribe to the efficacy of ablative techniques argue that no randomized studies exist to conclude that resection is better than ablation for CRLM, and that a combined resection/RFA approach to bCRLM results in less morbidity than two (or more) hepatic resections [1,13–15]. This clinical uncertainly is unfolding in an increasingly difficult healthcare climate. Many patients and practitioners continue to push for more health care at a time when healthcare costs, and specifically cancer care expenditures, are rising at an unsustainable rate [16,17]. These opposing forces require that clinicians, researchers and policy makers consider both cost and effectiveness in designing treatment strategies for patients [18,19]. Ultimately, for patients with bCRLM, the optimal treatment strategy has not been clearly defined. Randomized data do not exist to guide our
ß 2013 Wiley Periodicals, Inc.
clinical management, and for this reason cost must be included as a metric by which various treatment strategies are measured [19,20]. We hypothesized that because surgical intervention is a costly endeavor, 1‐ stage hepatic resection with RFA was a more cost‐effective strategy than 2‐stage hepatectomy for patients with bCRLM. Furthermore, we sought to determine if there were any peri‐operative or oncologic outcome differences.
MATERIALS AND METHODS Patient Selection The University of Texas MD Anderson Cancer Center Institutional Review Board approved a retrospective review of all patients with bCRLM who were treated with either 1‐stage (resection/RFA) or 2‐stage hepatectomy. Inclusion criteria were patients over 18 years of age with either metachronous or synchronous bCRLM. Exclusion criteria included patients who underwent any resection or ablation at an outside institution or who had their bCRLM resected in one setting. Patient‐specific variables including gender, location of primary tumor, metachronous versus synchronous disease, BMI, use of chemotherapy, size of hepatic lesions, extent of resection, resection margin, the presence of absence of PVE, repeat PVE, post‐operative morbidity/mortality,
The authors have no financial support to report. *Correspondence to: Daniel Abbott, MD, 234 Goodman St, ML 0772, Cincinnati, OH 45219. Fax: 513‐584‐0459. E‐mail: [email protected] Received 17 August 2013; Accepted 30 November 2013 DOI 10.1002/jso.23539 Published online 23 December 2013 in Wiley Online Library (wileyonlinelibrary.com).
Cost‐Effectiveness Liver Metastases local and/or distant recurrence, recurrence‐free survival and overall survival were collected and analyzed. RFA was performed according to previously published reports [10,21]. Briefly, all patients were treated during open laparotomy using ultrasound guided RF needle placement and lesions were ablated with the RF 2000 or RF 3000 generator system (Boston Scientific, Natick, MA). The treatment goal was always a 1 cm zone of normal parenchymal destruction around the tumor. Depending on lesion size, initial power was 50 W (3.5 cm array) or 80 W (4 cm array) and increased in 10 W increments at 1‐min intervals. For lesions >2.5 cm, the array was positioned multiple times.
Statistical Analysis Statistical analysis was performed with SAS Version 9.3 (Cary, NC). Continuous variables were compared using the Wilcoxon rank sum test. Categorical variables were compared using standard chi‐square analysis or Fisher’s exact test, as appropriate. Overall survival was calculated using the method of Kaplan and Meier, from the date of surgery (first surgery if 2‐stage approach) to the date of death. Recurrence‐free survival was calculated from the date of surgery to the date of cancer recurrence, either locoregional or systemic, or the date of death from another cause. A P‐value of
