- Email: [email protected]
A Multicenter Analysis of Distal Pancreatectomy for Adenocarcinoma: Is Laparoscopic Resection Appropriate?
A Multicenter Analysis of Distal Pancreatectomy for Adenocarcinoma: Is Laparoscopic Resection Appropriate? David A Kooby, MD, FACS, William G Hawkins, MD, FACS, C Max Schmidt, MD, PhD, FACS, Sharon M Weber, MD, FACS, David J Bentrem, MD, MS, FACS, Theresa W Gillespie, PhD, Johnita Byrd Sellers, MS, Nipun B Merchant, MD, FACS, Charles R Scoggins, MD, MBA, FACS, Robert CG Martin III, MD, PhD, FACS, Hong Jin Kim, MD, FACS, Syed Ahmad, MD, FACS, Clifford S Cho, MD, Alexander A Parikh, MD, FACS, Carrie K Chu, MD, Nicholas A Hamilton, MD, Courtney J Doyle, MD, Scott Pinchot, MD, Amanda Hayman, MD, MPH, Rebecca McClaine, MD, Attila Nakeeb, MD, FACS, Charles A Staley, MD, FACS, Kelly M McMasters, MD, PhD, FACS, Keith D Lillemoe, MD, FACS As compared with open distal pancreatectomy (ODP), laparoscopic distal pancreatectomy (LDP) affords improved perioperative outcomes. The role of LDP for patients with pancreatic ductal adenocarcinoma (PDAC) is not defined. STUDY DESIGN: Records from patients undergoing distal pancreatectomy (DP) for PDAC from 2000 to 2008 from 9 academic medical centers were reviewed. Short-term (node harvest and margin status) and longterm (survival) cancer outcomes were assessed. A 3:1 matched analysis was performed for ODP and LDP cases using age, American Society of Anesthesiologists (ASA) class, and tumor size. RESULTS: There were 212 patients who underwent DP for PDAC; 23 (11%) of these were approached laparoscopically. For all 212 patients, 56 (26%) had positive margins. The mean number of nodes (⫾ SD) examined was 12.6 ⫾8.4 and 114 patients (54%) had at least 1 positive node. Median overall survival was 16 months. In the matched analysis there were no significant differences in positive margin rates, number of nodes examined, number of patients with at least 1 positive node, or overall survival. Logistic regression for all 212 patients demonstrated that advanced age, larger tumors, positive margins, and node positive disease were independently associated with worse survival; however, method of resection (ODP vs. LDP) was not. Hospital stay was 2 days shorter in the matched comparison, which approached significance (LDP, 7.4 days vs. ODP, 9.4 days, p ⫽ 0.06). CONCLUSIONS: LDP provides similar short- and long-term oncologic outcomes as compared with OD, with potentially shorter hospital stay. These results suggest that LDP is an acceptable approach for resection of PDAC of the left pancreas in selected patients. (J Am Coll Surg 2010;210: 779–787. © 2010 by the American College of Surgeons) BACKGROUND:
Several investigators have compared outcomes for patients who underwent either laparoscopic or open distal (leftsided) pancreatectomy (DP).1-5 These reports consistently show that the laparoscopic approach provides patients with
improved perioperative outcomes over the open approach. The largest of these reports is the multicenter study from the Central Pancreas Consortium, in which patients who underwent the laparoscopic approach had lower intraoper-
Disclosure Information: Nothing to disclose. The Georgia Cancer Coalition provided financial support for the preparation of this article. Presented at Southern Surgical Association 121st Annual Meeting, Hot Springs, VA, December 2009.
(Hawkins, Hamilton), St Louis, MO; Indiana University (Schmidt, Doyle, Nakeeb, Lillemoe), Indianapolis, IN; University of Wisconsin (Weber, Cho, Pinchot), Madison, WI; Jesse Brown Veterans Affairs Medical Center (Bentrem, Hayman), Chicago, IL; Vanderbilt University (Merchant, Parikh), Nashville, TN; University of Louisville (Scoggins, Martin, McMasters), Louisville, KY; University of North Carolina Chapel Hill (Kim), Chapel Hill, NC; and University of Cincinnati (Ahmad, McClaine), Cincinnati, OH. Correspondence address: David A Kooby, MD, FACS, Dept. of Surgery, Emory University, Winship Cancer Institute, 1365C Clifton Rd, NE, 2nd Floor, Atlanta, GA 30322.
Received December 14, 2009; Accepted December 28, 2009. From the Departments of Surgery, Emory University School of Medicine (Kooby, Gillespie, Chu, Staley), Atlanta Veterans Affairs Medical Center (Kooby, Gillespie, Sellers, Staley), Atlanta, GA; Washington University
© 2010 by the American College of Surgeons Published by Elsevier Inc.
779
ISSN 1072-7515/10/$36.00 doi:10.1016/j.jamcollsurg.2009.12.033
780
Kooby et al
Laparoscopic Pancreatectomy for Cancer
Abbreviations and Acronyms
ASA BMI DP LDP ODP OR PDAC
⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽
American Society of Anesthesiologists classification body mass index distal pancreatectomy laparoscopic distal pancreatectomy open distal pancreatectomy odds ratio pancreatic ductal adenocarcinoma
ative blood loss, fewer postoperative complications, and shorter hospital stays than their open-approach counterparts in a matched cohort analysis.1 It can be concluded from these reports that provided proper patient selection and technique are used, the laparoscopic approach for leftsided pancreatectomy is comparable and perhaps better than the open approach for patients undergoing DP. Importantly, none of these reports focuses on and compares cancer outcomes (node retrieval, margin status, and survival) for the subset of patients diagnosed with adenocarcinoma of the pancreas. Although the Clinical Outcomes of Surgical Therapy study group trial proved that laparoscopic-assisted colectomy for colon cancer was not inferior to the traditional open approach,6 it is not clear if the results of this trial for colon cancer can be translated to a different, more aggressive malignancy such as pancreatic ductal adenocarcinoma (PDAC). There are few existing reports of long-term survival in patients undergoing laparoscopic distal pancreatectomy (LDP) for malignancies of the pancreas, and to date, no published report compares cancer outcomes among patients who underwent LDP versus open distal pancreatectomy (ODP). This study is a multicenter analysis of DP for patients with PDAC from the 9 university medical centers comprising the Central Pancreas Consortium. The goal of this study was to compare short-term (lymph node harvest and negative margin rate) and long-term (survival) oncologic outcomes in matched cohorts of patients undergoing either ODP or LDP for PDAC.
J Am Coll Surg
Cases were included if patients underwent DP (open or laparoscopic, with or without hand-access technique) for PDAC during the time period at one of the participating centers. Patients were excluded if they had adenocarcinoma with a background of intraductal papillary mucinous neoplasm or mucinous cystadenocarcinoma because behavior of these cancers differs from that of PDAC,7 and if there were insufficient demographic, operative, and outcomes data available for analysis and reporting. Definitions
American Society of Anesthesiologists (ASA) score was reported as stated in the operative record (range 1 to 4) and body mass index (BMI) was calculated for all patients as previously described.8 Greatest tumor diameter, as well as length of pancreas specimen resected (excluding the spleen), was recorded from the pathology report. Pancreas length was collected as a measure of extent of operation because it has been shown to be predictive of morbidity after left-sided pancreatectomy.9 Total node number represented the total number of lymph nodes identified and evaluated by the pathologist for each case, and positive node number corresponds to the total number of positive lymph nodes in a given patient. Margin positive resections included R1 (microscopically positive) and R2 (grossly positive) resections. Margin assessment also included radial tumor margins, ie, posterior pancreatic margin in addition to the pancreatic neck (or cut surface) margin of the gland, which is routinely assessed by most of the centers in the Central Pancreas Consortium. Adjuvant therapy was defined as use of preoperative or postoperative chemotherapy with or without radiation therapy. The focus of this project was not to examine the type or duration of chemotherapy, so these treatment strategies were combined. Furthermore, the participating centers comprising the Central Pancreas Consortium have similar philosophies on medical therapy for PDAC; patients with acceptable performance status are treated with chemotherapy in the postoperative setting, with radiation being added for positive surgical margins.10 Operative technique
METHODS This is an Institutional Review Board approved, multicenter retrospective review of prospectively maintained databases from 9 academic medical centers. These centers provided data as specified through a menu-driven database file, prepared with instructions for coding fields and dropdown lists where appropriate. A 28-field dataset was assessed, and only data pertaining to patients undergoing DP from January 1, 2000, through December 31, 2008, were included in this analysis.
Because this study pooled data from 9 university medical centers, no single technique was used for either ODP or LDP. Nonetheless, similar principles were applied, based on descriptions provided by participating centers. After appropriate diagnostic imaging (CT and/or MRI) and medical optimization, patients were routinely vaccinated against encapsulated bacteria and offered either ODP only or a choice of LDP or ODP on a case-by-case basis. After giving informed consent, patients were taken to the operating room using techniques previously described.11,12
Vol. 210, No. 5, May 2010
Kooby et al
Laparoscopic Pancreatectomy for Cancer
781
Table 1. Clinicopathologic Features of All Patients Undergoing Distal (Left) Pancreatectomy for Adenocarcinoma Variable
All (n ⴝ 212)
ODP (n ⴝ 189)
LDP (n ⴝ 23)
p Value*
Age, y Female, n (%) ASA, mean ⫾ SD BMI, mean ⫾ SD Operative time, min, mean ⫾ SD Blood loss, mL, mean ⫾ SD Tumor size, cm, mean ⫾ SD Total nodes, n, mean ⫾ SD Positive nodes, mean ⫾ SD Margin positive, n (%) Specimen length, cm, mean ⫾ SD Adjuvant therapy, n (%) Length of stay, d, mean ⫾ SD
65.4 ⫾ 11.4 120 (57) 2.6 ⫾ 0.7 26.4 ⫾ 6.0 231.3 ⫾ 79 750 ⫾ 804 4.4 ⫾ 2.7 12.6 ⫾ 8.4 1.3 ⫾ 1.9 56 (26) 9.8 ⫾ 3.0 146 (69) 10.0 ⫾ 7.0
65.5 ⫾ 11.3 109 (58) 2.6 ⫾ 0.7 26.2 ⫾ 6.0 230.4 ⫾ 80.4 790 ⫾ 828 4.5 ⫾ 2.8 12.5 ⫾ 8.5 1.4 ⫾ 1.9 50 (27) 9.9 ⫾ 2.9 133 (70) 10.7 ⫾ 6.3
65.1 ⫾ 12.3 11 (48) 2.7 ⫾ 0.7 28.5 ⫾ 5.7 238.4 ⫾ 68.1 422 ⫾ 473 3.5 ⫾ 1.3 13.8 ⫾ 8.4 1.0 ⫾ 1.8 6 (26) 9.4 ⫾ 3.7 13 (57) 7.4 ⫾ 3.4
NS 0.89 NS 0.38 NS 0.38 NS 0.38 NS 0.65 0.04 NS 0.10 NS 0.47 NS 0.36 NS 0.98 NS 0.47 NS 0.23 0.03
*p Value reflects comparison of ODP and LDP columns. ASA, American Society of Anesthesiologists classification; BMI, body mass index; LDP, laparoscopic distal pancreatectomy; NS, not significant; ODP, open distal pancreatectomy.
Converted cases were also included in the laparoscopic group as an intent-to-treat analysis.
variance between the groups was lowered and greater power was achieved.
Entire cohort and matched comparisons
Statistical analysis
Clinicopathologic parameters of ODP and LDP patients were compared in 2 ways: first, as an unmatched comparison of all patients to provide insight into which patients are being selected for the laparoscopic approach, and then as a matched cohort comparison to assess for outcomes differences between ODP and LDP, with minimized bias. Factors evaluated were patient demographics, including age, gender, and BMI; tumor factors, such as, tumor size and node status; and treatment factors, such as number of nodes assessed, length of pancreas resected, margin status, and use of adjuvant therapy. In the matched analysis, patients who underwent complex vascular resection and reconstruction were not included because none of these were performed in the LDP group. Three factors were used to select patients for the matched cohort comparison: age (years), ASA status (1 to 4), and tumor size (cm). Cases (LDP patients) were matched to controls (ODP patients) based on age within ⫾ 7 years, an ASA score within ⫾ 1, and a tumor size within ⫾ 0.5 cm. Separate lists of cases and controls were then generated and sorted by age, ASA class, and tumor size. The list of cases was chronologically examined and the 3 “best” available controls were selected. The results were double checked to make certain a control was used only once. In situations in which a control was used more than once, the control was removed and a second matching was performed. These steps were repeated until 3 ODP cases were uniquely matched to each LDP case. A 3:1 matching was chosen so that a larger set of patients could be captured. The ranges within each of the variables remained constant across the 2 groups. As a result, the
Results for patients who had ODP were compared with those who had LDP using chi-squared tests for categorical variables and Student’s t-tests for continuous variables. For the t-tests, equal variances were not assumed unless significant by Levene’s test. Two-tailed p values ⬍ 0.05 were considered statistically significant. The multivariate analyses for overall survival and margin-positive resection were performed using binary logistic regression. All continuous data fields were made dichotomous around the median value or according to accepted cutoff points in the literature. The results of the multivariate analysis were expressed as odds ratios (OR) with corresponding 95% confidence intervals. Survival analysis was performed using KaplanMeier estimates with log rank testing. All analyses were performed using SPSS 16.0 statistical software (SPSS Inc).
RESULTS A total of 1,755 distal pancreatectomies were captured from the 9 centers of the Central Pancreas Consortium during the study period. Of these, 228 (12.9%) were performed for PDAC, and 212 had sufficient data for analysis, representing the study population. Data for all 212 patients are shown in Table 1. Of these, 23 (10.8%) underwent LDP, and 189 (89.2%) had ODP. Of the 23 patients who had LDP, 4 (17%) were hand-access procedures and 4 (17%) were converted to open. Reasons for conversion were hemorrhage in 1 patient, failure to progress in 2 patients, and concern for margin adequacy in the fourth. Splenic preservation was performed in only 1 case, in which PDAC was not suspected.
782
Kooby et al
Laparoscopic Pancreatectomy for Cancer
J Am Coll Surg
Table 2. Multivariate Analysis for Factors Contributing to Worse Overall Survival for All 212 Patients Undergoing Distal Pancreatectomy for Pancreatic Ductal Adenocarcinoma Variable
Age ⬎ 65 y Gender (male) ASA ⬎ 2 BMI ⬎ 30 kg/m2 Operation (LDP) Blood loss ⬎ 500 mL Tumor size ⬎ 4 cm Margin positive Node positive Adjuvant therapy (none)
n
%
Overall survival p Value
119 92 131 53 23 126 110 56 112 66
56 43 62 25 11 59 52 26 53 31
0.03 0.10 0.37 0.30 0.75 0.52 0.01 0.02 ⬍0.01 ⬍0.01
Odds ratio
95% CI
1.59 1.39 1.20 1.26 1.11 1.15 1.66 1.70 1.92 2.08
1.05–2.39 0.94–2.04 0.83–1.81 0.82–1.95 0.60–2.03 0.74–1.80 1.1–2.49 1.10–2.63 1.24–2.95 1.30–3.32
ASA, American Society of Anesthesiologists classification; BMI, body mass index; LDP, laparoscopic distal pancreatectomy.
Patient demographics
Data for all 212 patients are shown in Table 1. Mean age for all patients was 65⫾11.4 years and 57% were women. The average operative time was 231 ⫾ 79 minutes, and blood loss for all patients was 750 ⫾ 804 mL. Average tumor size was 4.4 ⫾ 2.7 cm, average number of lymph nodes examined was 12.6 ⫾ 8.4, 54% had at least one positive lymph node, and 26% of patients had positive margins on final histologic analysis. Overall, 69% of patients received adjuvant therapy. Hospital length of stay for all 212 patients was 10.0 ⫾ 7.0 days. Table 1 also shows these factors broken down by surgical approach, and differences between the groups are noted. Patients undergoing LDP had significantly lower operative
blood loss (790 ⫾ 828 mL vs. 422 ⫾ 473 mL, p ⫽ 0.04), and hospital length of stay (10.7 ⫾ 7.3 days vs. 7.4 ⫾ 3.4 days, p ⫽ 0.03) as compared with ODP patients. Survival
At a median follow-up of 10 months, median actuarial overall survival for all 212 patients was 16 months (range 0 to 82 months). On multivariate analysis including all 212 patients, patient age greater than 65 years, tumor size greater than 4.0 cm, margin positive resection, node positive status, and lack of adjuvant therapy were independently associated with reduced overall survival. Method of resection (ODP vs. LDP) did not appear to affect survival, nor did gender, ASA ⬎2, BMI ⬎30 kg/m2, and blood loss ⬎500 mL (Table 2.). There were 2 recorded deaths within 30 days of operation (0.9%) and they were both in the ODP group (p ⫽ NS when compared with the laparoscopic group). Surgical margins
Figure 1. Overall survival for all patients undergoing either margin negative resection (solid line, n ⫽ 156, median survival 20 months) or margin positive resection (dotted line, n ⫽ 56, median survival 11 months) distal pancreatectomy for adenocarcinoma (p ⫽ 0.002, log rank).
Because resection technique potentially affects resection margin, we evaluated overall survival as a function of margin status (positive vs. negative) to determine which factors were associated with this outcome. As expected, the 56 patients (26%) with positive resection margins had reduced overall survival as compared with the 156 patients (74%) patients with negative margins (20 months vs. 11 months, p ⫽ 0.002, Fig. 1). On multivariate analysis including all 212 patients, only blood loss ⬎500 mL was independently associated with a positive margin resection. Patient age greater than 65 years, male gender, BMI ⬎ 30 kg/m2, ASA ⬎ 2, tumor size ⬎ 4.0 cm, and pancreatic specimen length were not associated with this finding. Importantly, method of resection (ODP vs LDP) did not have an impact on resection margin either (Table 3).
Vol. 210, No. 5, May 2010
Kooby et al
Laparoscopic Pancreatectomy for Cancer
783
Table 3. Multivariate Analysis for Factors Contributing to Margin Positive Resection for All 212 Patients Undergoing Distal Pancreatectomy for Pancreatic Ductal Adenocarcinoma Variable
Age ⬎ 65 y Gender (male) ASA ⬎ 2 BMI ⬎ 30 kg/m2 Operation (LDP) Blood loss ⬎ 500 mL Tumor size ⬎ 4 cm Pancreatic length ⬎10 cm
n
%
119 92 131 53 23 126 110 101
56 43 62 25 11 59 52 48
Overall survival p Value
0.20 0.14 0.53 0.27 0.75 0.01 0.08 0.60
Odds ratio
95% CI
1.55 1.63 1.20 1.25 1.36 2.59 1.82 1.19
0.79–3.02 0.86–3.11 0.63–2.46 0.72–3.33 0.46–4.00 1.24–5.43 0.94–3.53 0.62–2.27
ASA, American Society of Anesthesiologists classification; BMI, body mass index; LDP, laparoscopic distal pancreatectomy.
Matched analysis
To reduce the potential selection bias associated with comparing cohorts of patients undergoing open and minimally invasive surgery, a matched comparison was performed. We used a 3:1 ODP:LDP match controlling for age, ASA, and tumor size to include as many of the ODP patients as possible, given the relatively low number of LDP procedures performed in patients with PDAC. Of the 189 ODP patients, 70 were selected for matching. The cohorts had similar gender, length of operation, number of nodes evaluated, positive margin rates, pancreatic specimen length, and use of adjuvant therapy (Table 4). Patients undergoing LDP had a significantly higher BMI than those undergoing ODP. Other factors that approached significant differences between the groups included operative blood loss (329 mL higher in the ODP group, p ⫽ 0.08) and hospital Table 4. Clinicopathologic Features of the Matched Patients Undergoing Distal (Left) Pancreatectomy for Adenocarcinoma Variable
Age, y* Female ASA ⬎ 2* BMI, kg/m2 Operative time, min Blood loss, mL Tumor size, cm* Total nodes Positive nodes Margin positive Specimen length, cm Adjuvant therapy Length of stay, d
ODP (n ⴝ 70)
LDP (n ⴝ 23)
p Value
65.9 ⫾ 11.1 43 (61) 2.6 ⫾ 0.7 25.8 ⫾ 4.6 216 ⫾ 69 751 ⫾ 853 3.5 ⫾ 1.4 12.3 ⫾ 8.3 1.2 ⫾ 1.6 24 (34) 9.6 ⫾ 2.8 45 (64) 9.4 ⫾ 4.7
64.6 ⫾ 12.3 12 (48) 2.7 ⫾ 0.7 28.5 ⫾ 5.7 238 ⫾ 68 422 ⫾ 473 3.6 ⫾ 1.3 14.0 ⫾ 8.6 1.0 ⫾ 1.8 6 (26) 9.4 ⫾ 3.7 13 (57) 7.4 ⫾ 3.4
NS 0.76 NS 0.33 NS 0.29 0.03 NS 0.18 NS 0.08 NS 0.92 NS 0.41 NS 0.73 NS 0.61 NS 0.82 NS 0.62 NS 0.06
Results expressed as mean ⫾ standard deviation or n (%) where appropriate. *Variable used in matching process. ASA, American Society of Anesthesiologists classification; BMI, body mass index; LDP, laparoscopic distal pancreatectomy; NS, not significant; ODP, open distal pancreatectomy.
length of stay (2 days higher in the ODP group, p ⫽ 0.06). There was no difference in overall survival between the 2 treatment groups in this matched comparison (ODP, 16 months vs LDP 16 months, p ⫽ 0.71, Fig. 2).
DISCUSSION The aim of this study was to compare short-term (node harvest and margin status) and long-term (survival) oncologic outcomes in matched cohorts of patients undergoing either laparoscopic or open distal pancreatectomy for the diagnosis of pancreatic ductal adenocarcinoma. In this large, multicenter experience, patients undergoing LDP had similar oncologic outcomes as compared with those who had ODP in a matched cohort analysis, suggesting that a laparoscopic approach is acceptable for resection of selected patients with PDAC. Laparoscopic pancreatic resections are performed with increasing frequency, as evidenced by the growing number of publications in the field.12 LDP is more commonly performed than laparoscopic tumor enucleation or laparoscopic pancreaticoduodenectomy. Enucleations are not recommended for management of PDAC owing to potentially compromised tumor margins and absent nodal evaluations, and laparoscopic pancreaticoduodenectomy remains a technically challenging operation that is not yet well defined. So LDP serves as an appropriate model for exploring laparoscopic cancer outcomes in PDAC. There are limited reports of patients undergoing laparoscopic pancreatectomy for PDAC, and comparisons with open cohorts are not available. In 2005, Mabrut and colleagues13 reported an initial European multicenter feasibility experience with 127 patients who underwent laparoscopic pancreatic resection. In this study, only 4 patients had ductal adenocarcinoma. Although surgical margins were noted to be negative in these patients, limited information on other oncologic outcomes were reported.
784
Kooby et al
Laparoscopic Pancreatectomy for Cancer
Figure 2. Matched analysis overall survival for patients undergoing open (solid line, n ⫽ 70, median survival 16 months) versus laparoscopic (dotted line, n ⫽ 23, median survival 16 months) distal pancreatectomy for adenocarcinoma, in the 3:1 matched analysis (p ⫽ 0.71, log rank).
Dulucq and associates14 reported on 25 patients who underwent attempted laparoscopic pancreaticoduodenectomy, of which 11 cases were performed for PDAC. This report described clean surgical margins and absence of port-site metastases, but provided no survival outcomes for the subset of patients with PDAC. On average, 18 lymph nodes were examined for the PDAC cases, but no comparison was provided or reference made to average lymph node retrieval in open cases performed at the same time. One of the largest experiences with LDP for PDAC was reported by Fernandez-Cruz and coworkers.15 The authors described 13 cases with a 90% R0 resection rate, mean lymph node count of 14.5 ⫾ 3, and overall survival of 14 months. These data are compatible with ODP reports, but no direct comparison was provided because this paper focused on descriptive analysis. Any efforts to compare LDP and ODP cases in this setting would be limited by an inherent single-center institutional bias. In our report, the median number of nodes examined and positive margin rates were similar between LDP and ODP cases in the 3:1 matched comparison. Lymph node count is subject to pathologic analytic variability.16 Similarly, based on resection technique and method of pathologic assessment, margin status may vary as well.17 For this reason, our report provides a valuable comparison of 2 patient populations treated in a similar time period in a multi-institutional setting. Although the overall 26% positive margin rate seems high, it likely accurately represents a valid assessment of distal pancreatectomy for PDAC be-
J Am Coll Surg
cause the incidence of radial margin positivity is also routinely evaluated at the majority of the participating centers. Patient survival after resection for PDAC remains poor, with a median survival in the range of 17 to 27 months.10,18 Unfortunately, the results are worse for the subset of patients with left-sided primary tumors, presumably due to the nonspecific symptoms caused by these lesions, resulting in later stage of presentation as compared with right-sided lesions.19 As a result, fewer left-sided lesions are resectable at presentation. Although it was never the intent of our group to suggest LDP may affect cancer biology so to improve cancer-related survival over ODP, we performed this analysis to ensure these outcomes were not worse for our LDP patients. The overall survival of 16 months for the entire group of 212 patients is consistent with other reports in the literature. The subset of LDP patients had similar outcomes to the matched cohort of ODP patients. Although these are not randomized data, the results do suggest that the laparoscopic approach does not adversely affect overall survival in patients with PDAC, which is further demonstrated in the multivariate analysis. The factors that did have an adverse impact on patient survival are consistent with biologic factors that portend worse outcomes: advanced patient age, larger tumor size, lymph node and margin positive resections, and absence of adjuvant therapy. The margin positive resection rate in pancreatectomy for PDAC is difficult to interpret from existing series and randomized trials because there is no defined standard for histologic margin assessment. A recent review of the large randomized trials for resected PDAC illustrated this point; positive margin rates ranged from 0% to 83% depending on the trial.20 Reasons for this variability may include resection technique or pathologic assessment method.17,21 Existing reports of distal adenocarcinoma demonstrate resection positive margin rates of 17% to 28%.7,19,22 The findings of the current report are in line with those in the published literature, both with respect to margin positive rate and impact of margins on survival. Importantly, the laparoscopic approach does not appear to have resulted in a higher rate of positive margin resections. Hospital length of stay was significantly shorter for LDP versus ODP patients in the unmatched comparison, and it approached significance in the matched comparison. This finding is in keeping with findings in existing reports. In our initial Central Pancreas Consortium study of LDP versus ODP for all pancreatic diseases, hospital stay for LDP was approximately 1 day shorter than in this report.1 In our initial report, primarily benign pancreatic neoplasms were assessed. PDAC patients appear to have longer hospital stays, regardless
Vol. 210, No. 5, May 2010
Kooby et al
of method of resection.15 The strengths of this report are the multicenter experience and the variable experience with laparoscopic resection, allowing similar patients to be compared in both LDP and ODP groups, reducing selection bias. Furthermore, any single-center experience would have to span many more years to match the power provided by this consortium. The weaknesses of this study are the retrospective method and lack of surgical and histologic standardization.These are problems that can be addressed only by a prospective randomized trial, which would be difficult to accomplish given the infrequent diagnosis and operative suitability of patients with PDAC of the left pancreas. In conclusion, based on short- and long-term oncologic outcomes, this study supports the use of the laparoscopic approach for distal pancreatectomy in selected patients with ductal adenocarcinoma of the pancreas. This multicenter experience provides a balanced assessment because several of the centers included did not entertain the laparoscopic approach for patients with biopsy-proved or suspected PDAC of the left pancreas, making the matched comparison more valid. A randomized trial addressing this question is likely to fail because of the small target patient population that would satisfy criteria for enrollment.
5. Kim SC, Park KT, Hwang JW, et al. Comparative analysis of clinical outcomes for laparoscopic distal pancreatic resection and open distal pancreatic resection at a single institution. Surg Endosc 2008;22:2261–2268. 6. Clinical Outcomes of Surgical Therapy Study Group. A comparison of laparoscopically assisted and open colectomy for colon cancer.[see comment]. N Engl J Med 2004;350:2050–2059. 7. Christein JD, Kendrick ML, Iqbal CW, et al. Distal pancreatectomy for resectable adenocarcinoma of the body and tail of the pancreas. J Gastrointest Surg 2005;9:922–927. 8. Calle EE, Thun MJ, Petrelli JM, et al. Body-mass index and mortality in a prospective cohort of U.S. adults. N Engl J Med 1999;341:1097–1105. 9. Weber SH, Cho C, Merchant N, et al. Laparoscopic left pancreatectomy: complication risk score correlates with morbidity and risk for pancreatic fistula. Ann Surg Oncol 2009;16:2825–2833. 10. Merchant NB, Rymer J, Koehler EAS, et al. Adjuvant chemoradiation therapy for pancreatic adenocarcinoma: who really benefits? J Am Coll Surg 2009;208:829–838; discussion 838–841. 11. Kooby DA. Laparoscopic surgery for cancer: historical, theoretical, and technical considerations. Oncology 2006, 20:917–927; discussion 927–928, 931–932. 12. Kooby DA. Laparoscopic pancreatic resection for cancer. Expert Rev Anticancer Ther 2008;8:1597–1609. 13. Mabrut JY, Fernandez-Cruz L, Azagra JS, et al. Laparoscopic pancreatic resection: results of a multicenter European study of 127 patients. Surgery 2005;137:597–605. 14. Dulucq JL, Wintringer P, Mahajna A. Laparoscopic pancreaticoduodenectomy for benign and malignant diseases. Surg Endosc 2006;20:1045–1050. 15. Fernandez-Cruz L, Cosa R, Blanco L, et al. Curative laparoscopic resection for pancreatic neoplasms: a critical analysis from a single institution. J Gastrointest Surg 2007;11:1607–1621; discussion 1621–1622. 16. Adsay NV, Basturk O, Altinel D, et al. The number of nodes identified in a simple pancreaticoduodenectomy specimen: comparison of conventional vs orange-peeling approach in pathologic assessment. Mod Pathol 2009;22:107–112. 17. Strasberg SM, Linehan DC, Hawkins WG. Radical antegrade modular pancreatosplenectomy procedure for adenocarcinoma of the body and tail of the pancreas: ability to obtain negative tangential margins. J Am Coll Surg 2007;204:244–249. 18. Katz M, Wang H, Fleming J, et al. Long-term survival after multidisciplinary management of resected pancreatic adenocarcinoma. Ann Surg Oncol 2009;16:836–847. 19. Shoup M, Conlon KC, Klimstra D, Brennan MF. Is extended resection for adenocarcinoma of the body or tail of the pancreas justified? J Gastrointest Surg 2003;7:946–952; discussion 952. 20. Butturini G, Stocken DD, Wente MN, et al. Pancreatic cancer meta-analysis: influence of resection margins and treatment on survival in patients with pancreatic cancer: meta-analysis of randomized controlled trials. Arch Surg 2008;143:75–83; discussion 83. 21. Willett CG, Lewandrowski K, Warshaw AL, et al. Resection margins in carcinoma of the head of the pancreas. Implications for radiation therapy. Ann Surg 1993;217:144–148. 22. Neoptolemos JP, Stocken DD, Dunn JA, et al. Influence of resection margins on survival for patients with pancreatic cancer treated by adjuvant chemoradiation and/or chemotherapy in the ESPAC-1 randomized controlled trial.[see comment]. Ann Surg 2001;234:758–768.
Author Contributions Study conception and design: Kooby, Gillespie, Sellers Acquisition of data: Kooby, Kim, Chu, Hamilton, Doyle, Pinchot, Hayman, Merchant, Martin, McClaine Analysis and interpretation of data: Kooby, Gillespie, Sellers, Hawkins, Schmidt, Weber, Bentrem, Scoggins, Chu, Staley, McMasters, Lillemoe, Ahamd Drafting of manuscript: Kooby, Chu, Gillespie, Sellers, Hawkins, Schmidt, Weber, Bentrem, Scoggins, Lillemoe Critical revision: Kooby, Hawkins, Schmidt, Weber, Bentrem, Gillespie, Sellers, Merchant, Scoggins, Martin, Kim, Ahmad, Cho, Parikh, Chu, Hamilton, Doyle, Pinchot, Hayman, McClaine, Nakeeb, Staley, McMasters, Lillemoe
REFERENCES 1. Kooby DA, Gillespie T, Bentrem DJ, et al. Left-sided pancreatectomy: a multicenter comparison of laparoscopic and open approaches. Ann Surg 2008;248:438–446. 2. Velanovich V. Case-control comparison of laparoscopic versus open distal pancreatectomy. J Gastrointest Surg 2006;10:95–98. 3. Eom BW, Jang JY, Lee SE, et al. Clinical outcomes compared between laparoscopic and open distal pancreatectomy. Surg Endosc 2007;22:1334–1338. 4. The SH, Tseng D, Sheppard BC. Laparoscopic and open distal pancreatic resection for benign pancreatic disease. J Gastrointest Surg 2007;11:1120–1125.
Laparoscopic Pancreatectomy for Cancer
785
Several investigators have compared outcomes for patients who underwent either laparoscopic or open distal (leftsided) pancreatectomy (DP).1-5 These reports consistently show that the laparoscopic approach provides patients with
improved perioperative outcomes over the open approach. The largest of these reports is the multicenter study from the Central Pancreas Consortium, in which patients who underwent the laparoscopic approach had lower intraoper-
Disclosure Information: Nothing to disclose. The Georgia Cancer Coalition provided financial support for the preparation of this article. Presented at Southern Surgical Association 121st Annual Meeting, Hot Springs, VA, December 2009.
(Hawkins, Hamilton), St Louis, MO; Indiana University (Schmidt, Doyle, Nakeeb, Lillemoe), Indianapolis, IN; University of Wisconsin (Weber, Cho, Pinchot), Madison, WI; Jesse Brown Veterans Affairs Medical Center (Bentrem, Hayman), Chicago, IL; Vanderbilt University (Merchant, Parikh), Nashville, TN; University of Louisville (Scoggins, Martin, McMasters), Louisville, KY; University of North Carolina Chapel Hill (Kim), Chapel Hill, NC; and University of Cincinnati (Ahmad, McClaine), Cincinnati, OH. Correspondence address: David A Kooby, MD, FACS, Dept. of Surgery, Emory University, Winship Cancer Institute, 1365C Clifton Rd, NE, 2nd Floor, Atlanta, GA 30322.
Received December 14, 2009; Accepted December 28, 2009. From the Departments of Surgery, Emory University School of Medicine (Kooby, Gillespie, Chu, Staley), Atlanta Veterans Affairs Medical Center (Kooby, Gillespie, Sellers, Staley), Atlanta, GA; Washington University
© 2010 by the American College of Surgeons Published by Elsevier Inc.
779
ISSN 1072-7515/10/$36.00 doi:10.1016/j.jamcollsurg.2009.12.033
780
Kooby et al
Laparoscopic Pancreatectomy for Cancer
Abbreviations and Acronyms
ASA BMI DP LDP ODP OR PDAC
⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽
American Society of Anesthesiologists classification body mass index distal pancreatectomy laparoscopic distal pancreatectomy open distal pancreatectomy odds ratio pancreatic ductal adenocarcinoma
ative blood loss, fewer postoperative complications, and shorter hospital stays than their open-approach counterparts in a matched cohort analysis.1 It can be concluded from these reports that provided proper patient selection and technique are used, the laparoscopic approach for leftsided pancreatectomy is comparable and perhaps better than the open approach for patients undergoing DP. Importantly, none of these reports focuses on and compares cancer outcomes (node retrieval, margin status, and survival) for the subset of patients diagnosed with adenocarcinoma of the pancreas. Although the Clinical Outcomes of Surgical Therapy study group trial proved that laparoscopic-assisted colectomy for colon cancer was not inferior to the traditional open approach,6 it is not clear if the results of this trial for colon cancer can be translated to a different, more aggressive malignancy such as pancreatic ductal adenocarcinoma (PDAC). There are few existing reports of long-term survival in patients undergoing laparoscopic distal pancreatectomy (LDP) for malignancies of the pancreas, and to date, no published report compares cancer outcomes among patients who underwent LDP versus open distal pancreatectomy (ODP). This study is a multicenter analysis of DP for patients with PDAC from the 9 university medical centers comprising the Central Pancreas Consortium. The goal of this study was to compare short-term (lymph node harvest and negative margin rate) and long-term (survival) oncologic outcomes in matched cohorts of patients undergoing either ODP or LDP for PDAC.
J Am Coll Surg
Cases were included if patients underwent DP (open or laparoscopic, with or without hand-access technique) for PDAC during the time period at one of the participating centers. Patients were excluded if they had adenocarcinoma with a background of intraductal papillary mucinous neoplasm or mucinous cystadenocarcinoma because behavior of these cancers differs from that of PDAC,7 and if there were insufficient demographic, operative, and outcomes data available for analysis and reporting. Definitions
American Society of Anesthesiologists (ASA) score was reported as stated in the operative record (range 1 to 4) and body mass index (BMI) was calculated for all patients as previously described.8 Greatest tumor diameter, as well as length of pancreas specimen resected (excluding the spleen), was recorded from the pathology report. Pancreas length was collected as a measure of extent of operation because it has been shown to be predictive of morbidity after left-sided pancreatectomy.9 Total node number represented the total number of lymph nodes identified and evaluated by the pathologist for each case, and positive node number corresponds to the total number of positive lymph nodes in a given patient. Margin positive resections included R1 (microscopically positive) and R2 (grossly positive) resections. Margin assessment also included radial tumor margins, ie, posterior pancreatic margin in addition to the pancreatic neck (or cut surface) margin of the gland, which is routinely assessed by most of the centers in the Central Pancreas Consortium. Adjuvant therapy was defined as use of preoperative or postoperative chemotherapy with or without radiation therapy. The focus of this project was not to examine the type or duration of chemotherapy, so these treatment strategies were combined. Furthermore, the participating centers comprising the Central Pancreas Consortium have similar philosophies on medical therapy for PDAC; patients with acceptable performance status are treated with chemotherapy in the postoperative setting, with radiation being added for positive surgical margins.10 Operative technique
METHODS This is an Institutional Review Board approved, multicenter retrospective review of prospectively maintained databases from 9 academic medical centers. These centers provided data as specified through a menu-driven database file, prepared with instructions for coding fields and dropdown lists where appropriate. A 28-field dataset was assessed, and only data pertaining to patients undergoing DP from January 1, 2000, through December 31, 2008, were included in this analysis.
Because this study pooled data from 9 university medical centers, no single technique was used for either ODP or LDP. Nonetheless, similar principles were applied, based on descriptions provided by participating centers. After appropriate diagnostic imaging (CT and/or MRI) and medical optimization, patients were routinely vaccinated against encapsulated bacteria and offered either ODP only or a choice of LDP or ODP on a case-by-case basis. After giving informed consent, patients were taken to the operating room using techniques previously described.11,12
Vol. 210, No. 5, May 2010
Kooby et al
Laparoscopic Pancreatectomy for Cancer
781
Table 1. Clinicopathologic Features of All Patients Undergoing Distal (Left) Pancreatectomy for Adenocarcinoma Variable
All (n ⴝ 212)
ODP (n ⴝ 189)
LDP (n ⴝ 23)
p Value*
Age, y Female, n (%) ASA, mean ⫾ SD BMI, mean ⫾ SD Operative time, min, mean ⫾ SD Blood loss, mL, mean ⫾ SD Tumor size, cm, mean ⫾ SD Total nodes, n, mean ⫾ SD Positive nodes, mean ⫾ SD Margin positive, n (%) Specimen length, cm, mean ⫾ SD Adjuvant therapy, n (%) Length of stay, d, mean ⫾ SD
65.4 ⫾ 11.4 120 (57) 2.6 ⫾ 0.7 26.4 ⫾ 6.0 231.3 ⫾ 79 750 ⫾ 804 4.4 ⫾ 2.7 12.6 ⫾ 8.4 1.3 ⫾ 1.9 56 (26) 9.8 ⫾ 3.0 146 (69) 10.0 ⫾ 7.0
65.5 ⫾ 11.3 109 (58) 2.6 ⫾ 0.7 26.2 ⫾ 6.0 230.4 ⫾ 80.4 790 ⫾ 828 4.5 ⫾ 2.8 12.5 ⫾ 8.5 1.4 ⫾ 1.9 50 (27) 9.9 ⫾ 2.9 133 (70) 10.7 ⫾ 6.3
65.1 ⫾ 12.3 11 (48) 2.7 ⫾ 0.7 28.5 ⫾ 5.7 238.4 ⫾ 68.1 422 ⫾ 473 3.5 ⫾ 1.3 13.8 ⫾ 8.4 1.0 ⫾ 1.8 6 (26) 9.4 ⫾ 3.7 13 (57) 7.4 ⫾ 3.4
NS 0.89 NS 0.38 NS 0.38 NS 0.38 NS 0.65 0.04 NS 0.10 NS 0.47 NS 0.36 NS 0.98 NS 0.47 NS 0.23 0.03
*p Value reflects comparison of ODP and LDP columns. ASA, American Society of Anesthesiologists classification; BMI, body mass index; LDP, laparoscopic distal pancreatectomy; NS, not significant; ODP, open distal pancreatectomy.
Converted cases were also included in the laparoscopic group as an intent-to-treat analysis.
variance between the groups was lowered and greater power was achieved.
Entire cohort and matched comparisons
Statistical analysis
Clinicopathologic parameters of ODP and LDP patients were compared in 2 ways: first, as an unmatched comparison of all patients to provide insight into which patients are being selected for the laparoscopic approach, and then as a matched cohort comparison to assess for outcomes differences between ODP and LDP, with minimized bias. Factors evaluated were patient demographics, including age, gender, and BMI; tumor factors, such as, tumor size and node status; and treatment factors, such as number of nodes assessed, length of pancreas resected, margin status, and use of adjuvant therapy. In the matched analysis, patients who underwent complex vascular resection and reconstruction were not included because none of these were performed in the LDP group. Three factors were used to select patients for the matched cohort comparison: age (years), ASA status (1 to 4), and tumor size (cm). Cases (LDP patients) were matched to controls (ODP patients) based on age within ⫾ 7 years, an ASA score within ⫾ 1, and a tumor size within ⫾ 0.5 cm. Separate lists of cases and controls were then generated and sorted by age, ASA class, and tumor size. The list of cases was chronologically examined and the 3 “best” available controls were selected. The results were double checked to make certain a control was used only once. In situations in which a control was used more than once, the control was removed and a second matching was performed. These steps were repeated until 3 ODP cases were uniquely matched to each LDP case. A 3:1 matching was chosen so that a larger set of patients could be captured. The ranges within each of the variables remained constant across the 2 groups. As a result, the
Results for patients who had ODP were compared with those who had LDP using chi-squared tests for categorical variables and Student’s t-tests for continuous variables. For the t-tests, equal variances were not assumed unless significant by Levene’s test. Two-tailed p values ⬍ 0.05 were considered statistically significant. The multivariate analyses for overall survival and margin-positive resection were performed using binary logistic regression. All continuous data fields were made dichotomous around the median value or according to accepted cutoff points in the literature. The results of the multivariate analysis were expressed as odds ratios (OR) with corresponding 95% confidence intervals. Survival analysis was performed using KaplanMeier estimates with log rank testing. All analyses were performed using SPSS 16.0 statistical software (SPSS Inc).
RESULTS A total of 1,755 distal pancreatectomies were captured from the 9 centers of the Central Pancreas Consortium during the study period. Of these, 228 (12.9%) were performed for PDAC, and 212 had sufficient data for analysis, representing the study population. Data for all 212 patients are shown in Table 1. Of these, 23 (10.8%) underwent LDP, and 189 (89.2%) had ODP. Of the 23 patients who had LDP, 4 (17%) were hand-access procedures and 4 (17%) were converted to open. Reasons for conversion were hemorrhage in 1 patient, failure to progress in 2 patients, and concern for margin adequacy in the fourth. Splenic preservation was performed in only 1 case, in which PDAC was not suspected.
782
Kooby et al
Laparoscopic Pancreatectomy for Cancer
J Am Coll Surg
Table 2. Multivariate Analysis for Factors Contributing to Worse Overall Survival for All 212 Patients Undergoing Distal Pancreatectomy for Pancreatic Ductal Adenocarcinoma Variable
Age ⬎ 65 y Gender (male) ASA ⬎ 2 BMI ⬎ 30 kg/m2 Operation (LDP) Blood loss ⬎ 500 mL Tumor size ⬎ 4 cm Margin positive Node positive Adjuvant therapy (none)
n
%
Overall survival p Value
119 92 131 53 23 126 110 56 112 66
56 43 62 25 11 59 52 26 53 31
0.03 0.10 0.37 0.30 0.75 0.52 0.01 0.02 ⬍0.01 ⬍0.01
Odds ratio
95% CI
1.59 1.39 1.20 1.26 1.11 1.15 1.66 1.70 1.92 2.08
1.05–2.39 0.94–2.04 0.83–1.81 0.82–1.95 0.60–2.03 0.74–1.80 1.1–2.49 1.10–2.63 1.24–2.95 1.30–3.32
ASA, American Society of Anesthesiologists classification; BMI, body mass index; LDP, laparoscopic distal pancreatectomy.
Patient demographics
Data for all 212 patients are shown in Table 1. Mean age for all patients was 65⫾11.4 years and 57% were women. The average operative time was 231 ⫾ 79 minutes, and blood loss for all patients was 750 ⫾ 804 mL. Average tumor size was 4.4 ⫾ 2.7 cm, average number of lymph nodes examined was 12.6 ⫾ 8.4, 54% had at least one positive lymph node, and 26% of patients had positive margins on final histologic analysis. Overall, 69% of patients received adjuvant therapy. Hospital length of stay for all 212 patients was 10.0 ⫾ 7.0 days. Table 1 also shows these factors broken down by surgical approach, and differences between the groups are noted. Patients undergoing LDP had significantly lower operative
blood loss (790 ⫾ 828 mL vs. 422 ⫾ 473 mL, p ⫽ 0.04), and hospital length of stay (10.7 ⫾ 7.3 days vs. 7.4 ⫾ 3.4 days, p ⫽ 0.03) as compared with ODP patients. Survival
At a median follow-up of 10 months, median actuarial overall survival for all 212 patients was 16 months (range 0 to 82 months). On multivariate analysis including all 212 patients, patient age greater than 65 years, tumor size greater than 4.0 cm, margin positive resection, node positive status, and lack of adjuvant therapy were independently associated with reduced overall survival. Method of resection (ODP vs. LDP) did not appear to affect survival, nor did gender, ASA ⬎2, BMI ⬎30 kg/m2, and blood loss ⬎500 mL (Table 2.). There were 2 recorded deaths within 30 days of operation (0.9%) and they were both in the ODP group (p ⫽ NS when compared with the laparoscopic group). Surgical margins
Figure 1. Overall survival for all patients undergoing either margin negative resection (solid line, n ⫽ 156, median survival 20 months) or margin positive resection (dotted line, n ⫽ 56, median survival 11 months) distal pancreatectomy for adenocarcinoma (p ⫽ 0.002, log rank).
Because resection technique potentially affects resection margin, we evaluated overall survival as a function of margin status (positive vs. negative) to determine which factors were associated with this outcome. As expected, the 56 patients (26%) with positive resection margins had reduced overall survival as compared with the 156 patients (74%) patients with negative margins (20 months vs. 11 months, p ⫽ 0.002, Fig. 1). On multivariate analysis including all 212 patients, only blood loss ⬎500 mL was independently associated with a positive margin resection. Patient age greater than 65 years, male gender, BMI ⬎ 30 kg/m2, ASA ⬎ 2, tumor size ⬎ 4.0 cm, and pancreatic specimen length were not associated with this finding. Importantly, method of resection (ODP vs LDP) did not have an impact on resection margin either (Table 3).
Vol. 210, No. 5, May 2010
Kooby et al
Laparoscopic Pancreatectomy for Cancer
783
Table 3. Multivariate Analysis for Factors Contributing to Margin Positive Resection for All 212 Patients Undergoing Distal Pancreatectomy for Pancreatic Ductal Adenocarcinoma Variable
Age ⬎ 65 y Gender (male) ASA ⬎ 2 BMI ⬎ 30 kg/m2 Operation (LDP) Blood loss ⬎ 500 mL Tumor size ⬎ 4 cm Pancreatic length ⬎10 cm
n
%
119 92 131 53 23 126 110 101
56 43 62 25 11 59 52 48
Overall survival p Value
0.20 0.14 0.53 0.27 0.75 0.01 0.08 0.60
Odds ratio
95% CI
1.55 1.63 1.20 1.25 1.36 2.59 1.82 1.19
0.79–3.02 0.86–3.11 0.63–2.46 0.72–3.33 0.46–4.00 1.24–5.43 0.94–3.53 0.62–2.27
ASA, American Society of Anesthesiologists classification; BMI, body mass index; LDP, laparoscopic distal pancreatectomy.
Matched analysis
To reduce the potential selection bias associated with comparing cohorts of patients undergoing open and minimally invasive surgery, a matched comparison was performed. We used a 3:1 ODP:LDP match controlling for age, ASA, and tumor size to include as many of the ODP patients as possible, given the relatively low number of LDP procedures performed in patients with PDAC. Of the 189 ODP patients, 70 were selected for matching. The cohorts had similar gender, length of operation, number of nodes evaluated, positive margin rates, pancreatic specimen length, and use of adjuvant therapy (Table 4). Patients undergoing LDP had a significantly higher BMI than those undergoing ODP. Other factors that approached significant differences between the groups included operative blood loss (329 mL higher in the ODP group, p ⫽ 0.08) and hospital Table 4. Clinicopathologic Features of the Matched Patients Undergoing Distal (Left) Pancreatectomy for Adenocarcinoma Variable
Age, y* Female ASA ⬎ 2* BMI, kg/m2 Operative time, min Blood loss, mL Tumor size, cm* Total nodes Positive nodes Margin positive Specimen length, cm Adjuvant therapy Length of stay, d
ODP (n ⴝ 70)
LDP (n ⴝ 23)
p Value
65.9 ⫾ 11.1 43 (61) 2.6 ⫾ 0.7 25.8 ⫾ 4.6 216 ⫾ 69 751 ⫾ 853 3.5 ⫾ 1.4 12.3 ⫾ 8.3 1.2 ⫾ 1.6 24 (34) 9.6 ⫾ 2.8 45 (64) 9.4 ⫾ 4.7
64.6 ⫾ 12.3 12 (48) 2.7 ⫾ 0.7 28.5 ⫾ 5.7 238 ⫾ 68 422 ⫾ 473 3.6 ⫾ 1.3 14.0 ⫾ 8.6 1.0 ⫾ 1.8 6 (26) 9.4 ⫾ 3.7 13 (57) 7.4 ⫾ 3.4
NS 0.76 NS 0.33 NS 0.29 0.03 NS 0.18 NS 0.08 NS 0.92 NS 0.41 NS 0.73 NS 0.61 NS 0.82 NS 0.62 NS 0.06
Results expressed as mean ⫾ standard deviation or n (%) where appropriate. *Variable used in matching process. ASA, American Society of Anesthesiologists classification; BMI, body mass index; LDP, laparoscopic distal pancreatectomy; NS, not significant; ODP, open distal pancreatectomy.
length of stay (2 days higher in the ODP group, p ⫽ 0.06). There was no difference in overall survival between the 2 treatment groups in this matched comparison (ODP, 16 months vs LDP 16 months, p ⫽ 0.71, Fig. 2).
DISCUSSION The aim of this study was to compare short-term (node harvest and margin status) and long-term (survival) oncologic outcomes in matched cohorts of patients undergoing either laparoscopic or open distal pancreatectomy for the diagnosis of pancreatic ductal adenocarcinoma. In this large, multicenter experience, patients undergoing LDP had similar oncologic outcomes as compared with those who had ODP in a matched cohort analysis, suggesting that a laparoscopic approach is acceptable for resection of selected patients with PDAC. Laparoscopic pancreatic resections are performed with increasing frequency, as evidenced by the growing number of publications in the field.12 LDP is more commonly performed than laparoscopic tumor enucleation or laparoscopic pancreaticoduodenectomy. Enucleations are not recommended for management of PDAC owing to potentially compromised tumor margins and absent nodal evaluations, and laparoscopic pancreaticoduodenectomy remains a technically challenging operation that is not yet well defined. So LDP serves as an appropriate model for exploring laparoscopic cancer outcomes in PDAC. There are limited reports of patients undergoing laparoscopic pancreatectomy for PDAC, and comparisons with open cohorts are not available. In 2005, Mabrut and colleagues13 reported an initial European multicenter feasibility experience with 127 patients who underwent laparoscopic pancreatic resection. In this study, only 4 patients had ductal adenocarcinoma. Although surgical margins were noted to be negative in these patients, limited information on other oncologic outcomes were reported.
784
Kooby et al
Laparoscopic Pancreatectomy for Cancer
Figure 2. Matched analysis overall survival for patients undergoing open (solid line, n ⫽ 70, median survival 16 months) versus laparoscopic (dotted line, n ⫽ 23, median survival 16 months) distal pancreatectomy for adenocarcinoma, in the 3:1 matched analysis (p ⫽ 0.71, log rank).
Dulucq and associates14 reported on 25 patients who underwent attempted laparoscopic pancreaticoduodenectomy, of which 11 cases were performed for PDAC. This report described clean surgical margins and absence of port-site metastases, but provided no survival outcomes for the subset of patients with PDAC. On average, 18 lymph nodes were examined for the PDAC cases, but no comparison was provided or reference made to average lymph node retrieval in open cases performed at the same time. One of the largest experiences with LDP for PDAC was reported by Fernandez-Cruz and coworkers.15 The authors described 13 cases with a 90% R0 resection rate, mean lymph node count of 14.5 ⫾ 3, and overall survival of 14 months. These data are compatible with ODP reports, but no direct comparison was provided because this paper focused on descriptive analysis. Any efforts to compare LDP and ODP cases in this setting would be limited by an inherent single-center institutional bias. In our report, the median number of nodes examined and positive margin rates were similar between LDP and ODP cases in the 3:1 matched comparison. Lymph node count is subject to pathologic analytic variability.16 Similarly, based on resection technique and method of pathologic assessment, margin status may vary as well.17 For this reason, our report provides a valuable comparison of 2 patient populations treated in a similar time period in a multi-institutional setting. Although the overall 26% positive margin rate seems high, it likely accurately represents a valid assessment of distal pancreatectomy for PDAC be-
J Am Coll Surg
cause the incidence of radial margin positivity is also routinely evaluated at the majority of the participating centers. Patient survival after resection for PDAC remains poor, with a median survival in the range of 17 to 27 months.10,18 Unfortunately, the results are worse for the subset of patients with left-sided primary tumors, presumably due to the nonspecific symptoms caused by these lesions, resulting in later stage of presentation as compared with right-sided lesions.19 As a result, fewer left-sided lesions are resectable at presentation. Although it was never the intent of our group to suggest LDP may affect cancer biology so to improve cancer-related survival over ODP, we performed this analysis to ensure these outcomes were not worse for our LDP patients. The overall survival of 16 months for the entire group of 212 patients is consistent with other reports in the literature. The subset of LDP patients had similar outcomes to the matched cohort of ODP patients. Although these are not randomized data, the results do suggest that the laparoscopic approach does not adversely affect overall survival in patients with PDAC, which is further demonstrated in the multivariate analysis. The factors that did have an adverse impact on patient survival are consistent with biologic factors that portend worse outcomes: advanced patient age, larger tumor size, lymph node and margin positive resections, and absence of adjuvant therapy. The margin positive resection rate in pancreatectomy for PDAC is difficult to interpret from existing series and randomized trials because there is no defined standard for histologic margin assessment. A recent review of the large randomized trials for resected PDAC illustrated this point; positive margin rates ranged from 0% to 83% depending on the trial.20 Reasons for this variability may include resection technique or pathologic assessment method.17,21 Existing reports of distal adenocarcinoma demonstrate resection positive margin rates of 17% to 28%.7,19,22 The findings of the current report are in line with those in the published literature, both with respect to margin positive rate and impact of margins on survival. Importantly, the laparoscopic approach does not appear to have resulted in a higher rate of positive margin resections. Hospital length of stay was significantly shorter for LDP versus ODP patients in the unmatched comparison, and it approached significance in the matched comparison. This finding is in keeping with findings in existing reports. In our initial Central Pancreas Consortium study of LDP versus ODP for all pancreatic diseases, hospital stay for LDP was approximately 1 day shorter than in this report.1 In our initial report, primarily benign pancreatic neoplasms were assessed. PDAC patients appear to have longer hospital stays, regardless
Vol. 210, No. 5, May 2010
Kooby et al
of method of resection.15 The strengths of this report are the multicenter experience and the variable experience with laparoscopic resection, allowing similar patients to be compared in both LDP and ODP groups, reducing selection bias. Furthermore, any single-center experience would have to span many more years to match the power provided by this consortium. The weaknesses of this study are the retrospective method and lack of surgical and histologic standardization.These are problems that can be addressed only by a prospective randomized trial, which would be difficult to accomplish given the infrequent diagnosis and operative suitability of patients with PDAC of the left pancreas. In conclusion, based on short- and long-term oncologic outcomes, this study supports the use of the laparoscopic approach for distal pancreatectomy in selected patients with ductal adenocarcinoma of the pancreas. This multicenter experience provides a balanced assessment because several of the centers included did not entertain the laparoscopic approach for patients with biopsy-proved or suspected PDAC of the left pancreas, making the matched comparison more valid. A randomized trial addressing this question is likely to fail because of the small target patient population that would satisfy criteria for enrollment.
5. Kim SC, Park KT, Hwang JW, et al. Comparative analysis of clinical outcomes for laparoscopic distal pancreatic resection and open distal pancreatic resection at a single institution. Surg Endosc 2008;22:2261–2268. 6. Clinical Outcomes of Surgical Therapy Study Group. A comparison of laparoscopically assisted and open colectomy for colon cancer.[see comment]. N Engl J Med 2004;350:2050–2059. 7. Christein JD, Kendrick ML, Iqbal CW, et al. Distal pancreatectomy for resectable adenocarcinoma of the body and tail of the pancreas. J Gastrointest Surg 2005;9:922–927. 8. Calle EE, Thun MJ, Petrelli JM, et al. Body-mass index and mortality in a prospective cohort of U.S. adults. N Engl J Med 1999;341:1097–1105. 9. Weber SH, Cho C, Merchant N, et al. Laparoscopic left pancreatectomy: complication risk score correlates with morbidity and risk for pancreatic fistula. Ann Surg Oncol 2009;16:2825–2833. 10. Merchant NB, Rymer J, Koehler EAS, et al. Adjuvant chemoradiation therapy for pancreatic adenocarcinoma: who really benefits? J Am Coll Surg 2009;208:829–838; discussion 838–841. 11. Kooby DA. Laparoscopic surgery for cancer: historical, theoretical, and technical considerations. Oncology 2006, 20:917–927; discussion 927–928, 931–932. 12. Kooby DA. Laparoscopic pancreatic resection for cancer. Expert Rev Anticancer Ther 2008;8:1597–1609. 13. Mabrut JY, Fernandez-Cruz L, Azagra JS, et al. Laparoscopic pancreatic resection: results of a multicenter European study of 127 patients. Surgery 2005;137:597–605. 14. Dulucq JL, Wintringer P, Mahajna A. Laparoscopic pancreaticoduodenectomy for benign and malignant diseases. Surg Endosc 2006;20:1045–1050. 15. Fernandez-Cruz L, Cosa R, Blanco L, et al. Curative laparoscopic resection for pancreatic neoplasms: a critical analysis from a single institution. J Gastrointest Surg 2007;11:1607–1621; discussion 1621–1622. 16. Adsay NV, Basturk O, Altinel D, et al. The number of nodes identified in a simple pancreaticoduodenectomy specimen: comparison of conventional vs orange-peeling approach in pathologic assessment. Mod Pathol 2009;22:107–112. 17. Strasberg SM, Linehan DC, Hawkins WG. Radical antegrade modular pancreatosplenectomy procedure for adenocarcinoma of the body and tail of the pancreas: ability to obtain negative tangential margins. J Am Coll Surg 2007;204:244–249. 18. Katz M, Wang H, Fleming J, et al. Long-term survival after multidisciplinary management of resected pancreatic adenocarcinoma. Ann Surg Oncol 2009;16:836–847. 19. Shoup M, Conlon KC, Klimstra D, Brennan MF. Is extended resection for adenocarcinoma of the body or tail of the pancreas justified? J Gastrointest Surg 2003;7:946–952; discussion 952. 20. Butturini G, Stocken DD, Wente MN, et al. Pancreatic cancer meta-analysis: influence of resection margins and treatment on survival in patients with pancreatic cancer: meta-analysis of randomized controlled trials. Arch Surg 2008;143:75–83; discussion 83. 21. Willett CG, Lewandrowski K, Warshaw AL, et al. Resection margins in carcinoma of the head of the pancreas. Implications for radiation therapy. Ann Surg 1993;217:144–148. 22. Neoptolemos JP, Stocken DD, Dunn JA, et al. Influence of resection margins on survival for patients with pancreatic cancer treated by adjuvant chemoradiation and/or chemotherapy in the ESPAC-1 randomized controlled trial.[see comment]. Ann Surg 2001;234:758–768.
Author Contributions Study conception and design: Kooby, Gillespie, Sellers Acquisition of data: Kooby, Kim, Chu, Hamilton, Doyle, Pinchot, Hayman, Merchant, Martin, McClaine Analysis and interpretation of data: Kooby, Gillespie, Sellers, Hawkins, Schmidt, Weber, Bentrem, Scoggins, Chu, Staley, McMasters, Lillemoe, Ahamd Drafting of manuscript: Kooby, Chu, Gillespie, Sellers, Hawkins, Schmidt, Weber, Bentrem, Scoggins, Lillemoe Critical revision: Kooby, Hawkins, Schmidt, Weber, Bentrem, Gillespie, Sellers, Merchant, Scoggins, Martin, Kim, Ahmad, Cho, Parikh, Chu, Hamilton, Doyle, Pinchot, Hayman, McClaine, Nakeeb, Staley, McMasters, Lillemoe
REFERENCES 1. Kooby DA, Gillespie T, Bentrem DJ, et al. Left-sided pancreatectomy: a multicenter comparison of laparoscopic and open approaches. Ann Surg 2008;248:438–446. 2. Velanovich V. Case-control comparison of laparoscopic versus open distal pancreatectomy. J Gastrointest Surg 2006;10:95–98. 3. Eom BW, Jang JY, Lee SE, et al. Clinical outcomes compared between laparoscopic and open distal pancreatectomy. Surg Endosc 2007;22:1334–1338. 4. The SH, Tseng D, Sheppard BC. Laparoscopic and open distal pancreatic resection for benign pancreatic disease. J Gastrointest Surg 2007;11:1120–1125.
Laparoscopic Pancreatectomy for Cancer
785