44.11 Outcomes Associated with Esophagectomy: Results from a Large Prospectively Maintained Database

Posted on January 26, 2017

D. Lee1, P. Briceno2, R. Shridhar3, S. Kucera4, K. L. Meredith2  1Florida State University College Of Medicine,Sarasota, FL, USA 2Florida State University College Of Medicine/Sarasota Memorial Health Care System,Gastrointestinal Oncology,Sarasota, FL, USA 3University Of Central Florida,Radiation Oncology,Orlando, FL, USA 4Florida State University College Of Medicine/Sarasota Memorial Health Care System,Endoscopic Oncology,Sarasota, FL, USA

Introduction:  The long-term survival for patients with locally advanced esophageal cancer remains poor despite improvements in multi-modality care over the last several decades. Surgical resection remains piviotal in the management of patients with esophageal cancer.  We report our experience with esophageal cancer patients undergoing esophagectomy from a large prospectively maintained database. 

Methods:  A prospectively managed esophagectomy database was queried for patients undergoing esophagectomy 1996 and 2015. Basic demographics, tumor characteristics, operative details, and post-operative outcomes were recorded. Continuous variables were compared using the Kruskal Wallis or the ANOVA tests as appropriate. Pearson’s Chi-square test was used to compare categorical variables. All statistical tests were two-sided and an α (type I) error <0.05 was considered statistically significant. 

Results: We identified 856 patients who underwent esophagectomy with a mean age of 64 ± 10 years, mean BMI of 28.6 ± 6, and a mean follow up of 48 ± 13 months. There were 717 (83.8%) males and 139 (16.2%) females. Neoadjuvant therapy was administered in 543 (63.4%) patients. There were 475 (55.5 %) open Ivor Lewis (OIL), 69 (8.1%) open trans-hiatal (OTH), 10 (1.2%) open Mckeown, 95 (11.0%) minimally invasive esophagectomies (MIE) via Ivor Lewis approach (MIE IVL), 63 (7.4%) MIE TH and 144 (16.8%) robotic assisted Ivor Lewis esophagectomies (RAIL). There were 504 (58.8%) open esophagectomies and 302 (35.2%) MIE. Complications occurred less frequent in patients undergoing RAIL and MIE IVL:  145 (30.5%) OIL, 28 (40.6%) OTH, 28 (29.5%) MIE IVL, 31 (49.2%) MIE TH, and 34 (23.6%) RAIL (p=0.003). Pulmonary complications also occurred less frequently in RAIL and MIE IVL: 72 (15.2%) OIL, 12 (17.4%) OTH, 18 (28.6%) MIE TH, 8 (8.4%) MIE IVL, and 10 (6.9%) RAIL, p<0.001. Anastomotic leaks were less common in patients who underwent IVL either open or minimally invasive compared to trans-hiatal approaches: 23 (4.8%) OIL, 4 (4.2%) MIE IVL, 4 (2.8%) RAIL, versus 9 (13.0%) OTH, 4 (6.3%) MIE TH, p=0.03. There were 13 (1.5%) mortalities and this did not differ among techniques, p= 0.6. Oncologic quality as indicated by R0 resections and mean lymph node harvest were improved in patients undergoing RAIL: 449 (94.7%) and10±6 OIL, 62 (89.9%) and 8±5 OTH, 60 (96.8%) and 9±6 MIE TH, 86 (93.5%) and 14±7 MIE IVL, and 144 (100%) and 20±9 RAIL, p=0.04 and p=0.001. Median length of hospitalization was 9 days in both RAIL and MIE IVL and 10 days in all other groups, p=0.2. 

Conclusion: We report our experience with varying approaches to esophageal resections from a large esophagectomy database.  Minimally invasive and robotic Ivor Lewis techniques demonstrated substantial benefits in post-operative complications.  Oncologic outcomes similarly favor MIE IVL and RAIL.  Pulmonary outcomes were not reduced by trans-hiatal approaches.

 

44.10 High Perioperative Morbidity and Mortality in Patients with Malignant Nonfunctional Adrenal Tumors

Posted on January 26, 2017

A. R. Marcadis1, G. A. Rubio1, Z. F. Khan1, J. C. Farra1, T. M. Vaghaiwalla1, J. I. Lew1  1University Of Miami,Leonard M. Miller School Of Medicine, Division Of Endocrine Surgery, DeWitt Daughtry Family Department Of Surgery,Miami, FL, USA

Introduction: Adrenal gland tumors are categorized into those that produce excess hormones (functional) and those that do not (nonfunctional). Both functional and nonfunctional adrenal tumors can be further subdivided by benign and malignant pathology. Malignant nonfunctional adrenal tumors are rare, with definitive diagnosis often made by final pathology. Furthermore, the morbidity and mortality associated with surgical treatment of such uncommon tumors remains uncertain. This study compares the perioperative in-hospital outcomes after adrenalectomy in patients with benign and malignant nonfunctional adrenal tumors.

Methods: A retrospective cross-sectional analysis was performed using the Nationwide Inpatient Sample database (2006-2011) to identify surgical patients hospitalized for adrenal tumors. Patients who underwent unilateral open or laparoscopic adrenalectomy for nonfunctional adrenal tumors were further subdivided into benign and malignant groups based on final pathology. Patient demographics, socioeconomic and clinical factors, comorbidities, and perioperative complications were evaluated by univariate and risk-adjusted multivariate logistic regression. Data were analyzed using two-tailed Chi-square and t-tests.

Results: Of 28,339 patients who underwent unilateral adrenalectomy for nonfunctional adrenal tumors, 75% (n=21,279) had benign adenomas, while the remaining 25% (n=7,060) had malignancy on final pathology. Patients with malignant nonfunctional adrenal tumors were more likely to be younger in age (46 vs 54 years; p<0.01) and men (61.6 vs 44.6%; p<0.01) compared to patients with benign nonfunctional adrenal tumors. Patients with malignant nonfunctional adrenal tumors were more likely to suffer intraoperative complications including vascular (7.1 vs 3.8%; p<0.01) and splenic injury (5.6 vs 2.1%; p<0.01), postoperative complications including hematoma (3.6 vs 2.0%; p<0.01), shock (1.1 vs 0.5%; p<0.01), acute kidney injury (3.1 vs 2.5%; p<0.01), venous thromboembolism (1.5 vs 0.6%; p<0.01), and pneumothorax (1.6 vs 0.9%; p<0.01), as well as have higher rates of blood transfusion (18.5 vs 7.0%; p<0.01), longer hospital stay (5.9 vs 4.2 days; p<0.01) and higher hospital charges (59,529 vs $45,152; p<0.01) compared to their benign counterparts. Finally, patients with malignant nonfunctional adrenal tumors had a significantly higher in-hospital mortality compared to patients with benign nonfunctional adrenal tumors (1.0 vs 0.4%; p <0.01).

Conclusion: Patients with malignant nonfunctional adrenal tumors have significantly higher perioperative morbidity and mortality compared to their benign nonfunctional counterparts. Men younger than 50 years of age with nonfunctional adrenal tumors have an increased risk for underlying malignancy. Such patients should be counseled and medically optimized in preparation for adrenalectomy, and surgeons should remain vigilant in preventing perioperative complications.

44.09 Adjuvant Radiation Provides Limited Survival Benefit After R1 Resections For Pancreatic Head Cancer

Posted on January 26, 2017

N. R. Suss1, M. S. Talamonti1,2, D. S. Bryan2, C. Wang1, K. M. Kuchta1, S. J. Stocker1, D. J. Bentrem3, K. K. Roggin2, D. J. Winchester1,2, R. Marsh1,2, R. A. Prinz1,2, M. S. Baker1,2  1NorthShore University HealthSystem,Department Of Surgery,Evanston, IL, USA 2University Of Chicago,Department Of Surgery,Chicago, IL, USA 3Feinberg School Of Medicine,Department Of Surgery,Chicago, IL, USA

Introduction: The benefit of adding radiation to adjuvant systemic chemotherapy in patients that have undergone a margin positive resection for early stage pancreatic cancer (PDAC) has not been well established. 

Methods: We queried the National Cancer Database (NCDB) for 2004 through 2012 to identify patients with pathologic stage I-II PDAC of the pancreatic head who underwent pancreaticoduodenectomy and had a microscopic positive margin on final pathology (R1 resection). Kaplan-Meier, multivariable and cox regression modeling were employed to identify factors associated with radiation use and compare overall survival for patients receiving adjuvant chemotherapy with radiation (CRT) to those receiving adjuvant chemotherapy alone (ACT).  Patients receiving neoadjuvant therapy and those who did not receive adjuvant chemotherapy were excluded.

Results: 1,310 patients met inclusion criteria. 255 (19.5%) were lymph node negative (Stages IA, IB, IIA) and 1,055 (80.5%) were node positive (Stage IIB). 897 (68.5%) patients received CRT, while 413 (31.5%) received ACT. Multivariable stepwise logistic regression identified younger age (OR 2.310, 95% CI [1.515, 3.521]), treatment in New England (OR 7.915, 95% CI [3.369,18.595]), and negative nodal status (OR 1.797, 95% CI [1.286, 2.511]) as independently associated with use of CRT.  Cox modeling adjusting for age, sex, race, comorbid disease state, socioeconomic status (SES), insurance status, facility type and volume, surgery type, vascular abutment, pathological T stage, and nodal status, identified High SES (HR 0.717, 95% CI [0.609, 0.846]) and use of CRT (HR 0.828 95% CI [0.726, 0.944]) as independently associated with improved overall survival. Charlson score of one (HR  1.226 95% CI [1.069-1.406]) and node positivity (HR 1.547 95% CI [1.317, 1.817]) were independently associated with higher risk of mortality. Cox modeling stratified by stage demonstrated the benefit of radiation to be statistically significant in node positive patients only.  Node positive patients undergoing CRT demonstrated a median survival of 17.1 months vs. 14.8 months for node positive patients undergoing ACT (p=0.001). In patients who were lymph node negative, there was no difference in overall survival with radiation (21.9 vs. 23.3 months, p=0.457) [Figure 1].

Conclusion: Addition of radiation to adjuvant chemotherapy confers a limited survival benefit over treatment with chemotherapy alone in patients having an R1 resection for lymph node positive pancreatic head cancer.  Radiation offers no benefit for patients undergoing an R1 resection for disease that is node negative. Randomized trials are needed to better identify subgroups of PDAC patients for whom benefits of radiation justify the known risks. 

 

44.08 Impact of Neoadjuvant Radiation and Chemotherapy on Perioperative Complications Following Whipple

Posted on January 26, 2017

T. Tan1, C. McDaniel1, W. W. Zhang1, D. Rybin2, G. Doros2, Q. Chu1  1Louisiana State University Health Sciences Center,Depatment Of Surgery,Shreveport, LA, USA 2Boston Medical Center,Boston, MA, USA

Introduction: We evaluated the outcomes of patients undergoing pancreaticoduodenectomy who received neoadjuvant radiation therapy and chemotherapy for pancreatic cancer.

 

Methods: Using the National Surgical Quality Improvement Program dataset (2005-2012), we identified and examined 10,217 pancreaticoduodenectomy using the ICD-9 and CPT codes. Patients were divided into two groups based on whether they received neoadjuvant radiation and chemotherapy. Outcomes evaluated were perioperative mortality and morbidity. Multivariable logistic regression was used to examine association between neoadjuvant therapy and perioperative outcomes adjusting for possible confounders.

 

Results: There were 10,217 pancreaticoduodenectomys in this study cohort with 488 patients (5%) receiving neoadjuvant therapy prior to surgery. Patients who were treated with neoadjuvant therapy had significant higher history of >10% weight lost prior to surgery (26% vs. 18%, p<.001), chronic steroid therapy (3% vs. 2%, p=.03), and significantly longer operative time (443±140 vs. 373±128 minutes, p<.001). Although perioperative survival was similar between two cohorts, those treated with neoadjuvant therapy had significantly higher risk of surgical site infection (SSI) (14% vs. 10%, p=.002), thromboembolism (5% vs. 3%, p=.03) but lower risk of pneumonia (3% vs. 5%, p=.10).  In multivariable analysis, neoadjuvant therapy was associated with increased risk of SSI (OR 1.4, 95% CI 1, 1.8, p=.02), but lower risk of pneumonia (OR 0.5, 95% CI 0.3, 0.8, p=.007).

Conclusion: Patients who received neoadjuvant radiation therapy and chemotherapy have significant higher risk of surgical site infection following pancreaticoduodenectomy for pancreatic cancer. Further studies are required to evaluate appropriate role of neoadjuvant therapy in patients undergoing surgical treatment for pancreatic cancer. 

44.07 Complete Response to Neoadjuvant Chemoradiation Does Not Increase Morbidity After Esophagectomy.

Posted on January 26, 2017

A. M. Brown1, D. Giugliano1, F. Palazzo1, E. L. Rosato1, N. R. Evans1, C. R. Lamb1, D. A. Levine1, A. C. Berger1  1Thomas Jefferson University,Department Of Surgery,Philadelphia, PA, USA

Introduction:
Neoadjuvant chemoradiation (nCRT) followed by definitive esophagectomy has become a mainstay of treatment for stage two and three esophageal cancer. It has been demonstrated in rectal cancer that a complete response to nCRT is a predictor of anastomotic leak, as well as post-operative morbidity and mortality. We hypothesized that a complete treatment effect after nCRT may negatively affect anastomotic leak rate and post-operative morbidity and mortality.

Methods:
A retrospective review of all patients who underwent esophagectomy following nCRT for esophageal cancer between January 2000, and June 2016 was completed. The patients were stratified by their response to preoperative chemoradiation: no change or upstage (group 1), partial response (group 2), or a complete response (group 3), based on final surgical pathology. The postoperative courses of all patients were reviewed for anastomotic leak, respiratory failure defined as ventilator dependence greater than 48 hours, re-intubation, or acute respiratory distress syndrome (ARDS), as well as any pulmonary complication. All complications were categorized using a modification of the Clavien-Dindo classification. Statistical significance was calculated using a one-sided ANOVA test. 

Results:
There were 215 patients who underwent esophagectomy for esophageal cancer. The average age was 61 (range 31 to 84). Open esophagectomy was performed in 91 patients, and 124 underwent minimally invasive esophagectomy. With regards to neoadjuvant treatment, 78 patients (36%) had no change or a pathological upstage, 69 (32%) had a partial response, and 68 (32%) had a complete response. Rates of anastomotic leak were 14.1% in group 1, 8.7% in group 2, and 17.6% in group 3 (p=0.306). Rates of respiratory failure were similar—21.8%, 23.2%, and 23.5% respectively (p=0.965). Grade 3 or higher complication rates were 29.5%, 31.9%, and 30.9% respectively (p=0.952). There were no major differences in the frequency of any pulmonary complication (41.0%, 43.5%, 35.3%, p=0.607), or peri-operative mortality (7.7%, 5.8%, and 7.4% (p=0.895)). 

Conclusion:
There are no significant differences in complications and anastomotic leak based on pathologic response to nCRT.  Esophagectomy after nCRT is not compromised by a complete pathologic response.
 

44.06 Prognostic Nomogram for Patients with Operable Pancreatic Cancer Treated with Neoadjuvant Therapy

Posted on January 26, 2017

S. Jeong1, M. Aldakkak1, K. Ahn2, C. Huang3, K. K. Christians1, B. A. Erickson4, P. S. Ritch5, B. George5, D. B. Evans1, S. Tsai1  4Medical College Of Wisconsin,Radiation Oncology,Milwaukee, WI, USA 5Medical College Of Wisconsin,Hematology Oncology/Dept Of Medicine,Milwaukee, WI, USA 1Medical College Of Wisconsin,Surgery,Milwaukee, WI, USA 2Medical College Of Wisconsin,Biostatistics,Milwaukee, WI, USA 3University Of Wisconsin, Milwaukee,Biostatistics/Joseph J. Zilber School Of Public Health,Milwaukee, WI, USA

Introduction: American Joint Committee on Cancer (AJCC) TNM staging system provides population based estimates of survival based on pathologic variables.  For patients with pancreatic cancer (PC), survival estimates were generated from patients who have undergone surgery with or without postoperative therapy. Increasingly, preoperative (neoadjuvant) therapy followed by surgery is being utilized in patients with PC in recognition of the high risk of disease recurrence and the inability to consistently deliver adjuvant therapy after pancreatectomy. Whether the AJCC staging system accurately predicts survival among patients who have received neoadjuvant therapy is unclear. We assessed the survival discrimination of the AJCC staging system for patients with PC who have received neoadjuvant therapy and surgery, and developed a novel prognostic nomogram using clinical variables.

Methods: Clinical data and survival outcomes from patients with PC who completed neoadjuvant therapy and surgery at a single institution were collected. Survival at 1-, 2-, and 3-years from the date of restaging after neoadjuvant therapy and surgery were used for the purpose of nomogram construction. Concordance index (c-index) and calibration plots were used to assess predictive accuracy. The nomogram was developed using multivariable Cox proportional hazards model. Clinical stage was defined at the time of diagnosis and patients were categorized as having resectable or borderline resectable disease. Postoperative carbohydrate antigen 19-9 (CA19-9) was measured at the first restaging visit after surgery.

Results: Data was available for 167 patients with resectable and borderline resectable PC. The concordance indices (c-index) for 1-, 2-, and 3- years using the AJCC staging system were 0.57, 0.55, and 0.56, respectively. Clinical stage (HR:2.31; 95%CI:1.48-3.63) and postoperative CA19-9 levels (HR: 2.14; 95%CI:1.38-3.34) were the strongest prognostic factors. A parsimonious nomogram (Figure 1) including clinical stage, postoperative CA19-9, and age predicted 1-, 2-, and 3- year survival with c-indices of 0.66, 0.66, and 0.66, respectively. Calibration plots showed good fitness between observed and predicted probabilities. A combined nomogram using clinical stage, postoperative CA19-9, age, and AJCC stage demonstrated c-indices for 1-, 2-, and 3-years of 0.68, 0.67, and 0.67, respectively.

Conclusion: AJCC staging system poorly discriminates survival for patients who have received neoadjuvant therapy and surgery. A prognostic nomogram utilizing clinical stage and postoperative CA19-9 levels provides more accurate survival prediction than the AJCC model.  External validation will be performed to assess the generalizability of the nomogram. 

44.05 Overall Survival after Resection for Retroperitoneal Sarcoma at Academic vs. Community Centers

Posted on January 26, 2017

N. G. Berger1, J. P. Silva1, K. K. Christians1, S. Tsai1, T. Gamblin1  1Medical College Of Wisconsin,Surgery, Division Of Surgical Oncology,Milwaukee, WI, USA

Introduction:  Surgical resection remains the curative therapy for retroperitoneal sarcoma (RPS). Data recently published shows a correlation between improved outcomes for complex oncologic operations and treatment at academic centers. For large retroperitoneal sarcomas, surgical resection can be complex and require multidisciplinary care. The present study hypothesized that survival rates vary between treating center for patients undergoing resection for retroperitoneal sarcoma.

Methods:  Patients with all-stage and all-size retroperitoneal sarcomas who underwent surgical resection were identified from the National Cancer Database (2004-2013). Treating centers were categorized as Academic Cancer Centers (ACC) or Community Cancer Centers (CCC). OS was analyzed by log-rank test and graphed using Kaplan-Meier method.

Results: A total of 5,106 patients were identified. Median age of diagnosis was 63. The majority of patients (58.2%, n=2,970) patients underwent resection at an ACC.  Improved median OS was seen at ACCs across all stages compared to CCCs (79.1 months vs. 64.3 months; p=0.004). ACCs exhibited a higher rate of R0 resection (51.2% vs. 44.1%, p<0.001). No difference between cohorts was seen for 90-day mortality or 30-day readmission rates, though 30-day mortality at CCCs following resection did trend toward significance (1.9% vs. 2.9%, p=0.061). On Cox univariate regression, age, sex, ethnicity, Charlson Comorbidity Score, tumor size, tumor grade, and treating facility were identified as significant factors. Following multivariate regression, treatment at an academic center was associated with a significant hazard ratio (HR) for survival (HR= 0.91, IQR 0.82-1.00, p=0.045).

Conclusion: Resection for RPS performed at ACC is associated with improved survival compared to CCC, and an improved HR for survival on multivariate regression. This suggests that site of care plays a role in patient outcomes.

 

44.04 Near infrared perfusion assessment of gastric conduit in minimally invasive Ivor Lewis esophagectomy

Posted on January 26, 2017

B. G. Dalton1, A. A. Ali2, Z. T. Awad1  1University Of Florida-Jacksonville,Surgery,Jacksonville, FL, USA 2Wayne State University,Surgery,Detroit, MI, USA

Introduction:

Anastomotic leak and conduit necrosis can be devastating complications following Ivor Lewis esophagectomy. Conduit ischemia is the leading cause of graft necrosis and anastomotic leak. Near infrared imaging (NIR) using IndoCyanine Green (ICG) allows for real time assessment of tissue perfusion.  We theorized that the use of intraoperative NIR during laparoscopic minimally invasive Ivor Lewis esophagectomy (MIE) would allow for resection of a greater portion of gastric conduit which may reduce leak rate.

Methods:

After IRB approval retrospective analysis of a prospectively collected data from 2014-2015 of 40 consecutive MIE was performed.  All operations were performed for esophageal cancer by a single surgeon at a tertiary referral center. Intravenous ICG and laparoscopic NIR (Pinpoint, Novadaq, Canada) were used to visualize and assess gastric conduit perfusion for the most recent 20 patients in the study (NIR group). Extended conduit resection was performed if ischemia was present on NIR.  The non-NIR group was composed of the 20 MIE cases immediately prior to the advent of NIR use in our practice. Comparative analysis was performed using student t test for continuous variables and Fishers exact for binary variables.  Statistical significance is defined as P ≤ 0.05.

Results:

No differences were found between the 2 groups with regard to age, gender, BMI, clinical stage, pathologic stage, or comorbidities.  Comparing non-NIR to NIR groups, no statistically significant differences were found in overall complication rate (55% vs 40%, p=0.53), reoperation within the same admission (5% vs 10%, p=1), 90 day readmission (10% vs 10%, p=1) and 90 day reoperation (10% vs 5%, p=1). NIR resulted in extended level of proximal conduit resection in 30% (6/20) in the NIR group. Two patients in NIR group developed anastomotic leak (2/20) while no patients in the non-NIR group were found to have leaks (p=0.48). Both leaks were in patients that had additional conduit resection after NIR technology was used to assess conduit perfusion. Endoscopic stent placement was used to manage both leaks, and operative drainage or repair were not required.  One mortality related to graft necrosis was noted in the non-NIR group, while there were 0 mortalities in the NIR group. (p=1.0).

Conclusion:

Although near infrared angiography plays a vital role in assessment of tissue perfusion, in our study its use did not result in reduction of anastomotic leak rate. However, this technology did allow for additional resection of ischemic portions of the gastric conduit.  This extended resection potentially prevented extensive conduit necrosis.  Larger studies are needed to validate the use of this novel technology.

44.03 Pilot Study of SAVI SCOUT® to Localize Non-Palpable Breast Lesions to Reduce Re-excision

Posted on January 26, 2017

R. Shirley1, P. Peddi1, S. Ahmed1, Q. D. Chu1  1Louisiana State University Health Sciences Center-Shreveport,Surgery,Shreveport, LA, USA

Introduction: Wire localization (WL) is standard preoperative procedure to localize non-palpable breast lesions.  The SAVI SCOUT® guidance system is an FDA-approved medical device that uses non-radioactive electromagnetic wave technology and serves as an alternate to WL technique.  The purpose of the study is to compare the re-excision rates between WL and SAVI SCOUT® and assess the ease of performing such a technique by surgical trainees.   

Methods: We performed an IRB approved retrospective chart review of all women undergoing WL from 2011-2015 and compared them to women undergoing SAVI SCOUT® technique from 2015-2016. Re-excision rates, weights of the final specimens, and rate of detection by surgical trainees were calculated. Statistical t-tests and chi square tests were used.  P-value ≤0.05 was considered as statistically significant. 

Results:Of the 116 WL breast cancer biopsies performed, 43 required re-excision (37%).  Of the 26 SAVI SCOUT® performed, 17 were malignant; of these 17, only 2 required re-excision (11.8%; P=0.04).  This translates to a reduction of 68.2% with SAVI SCOUT®.  The average specimen weight for the WL group was 63g versus 55g for the SAVI SCOUT® group (P=0.38.  The average margin width was 2.7 mm for the WL versus 3.0mm for the SAVI SCOUT® (P=0.43). Surgical residents were successful in localizing the lesions in 25 out of 26 (96%) patients using SAVI SCOUT® technique.   

Conclusion:The re-excision rate was significantly lower with SAVI SCOUT® and can easily be done by surgical residents. Given its advantages, SAVI SCOUT® should be considered over WL technique.  

 

44.02 Risk Factors for Post-operative Complications After Lymph Node Dissection for Melanoma

Posted on January 26, 2017

B. Sunkara1, S. Diljak1, R. D. Kramer1, R. J. Strobel1, D. J. Mercante1, J. S. Jehnsen1, J. F. Friedman1, A. Durham1, T. Johnson1, M. S. Cohen1  1University Of Michigan,Department Of Surgery,Ann Arbor, MI, USA

Introduction:  While the literature looking at complications after lymphadenectomy for melanoma is fairly robust, the root cause analysis of risk factors contributing to those complications has not been as well studied and often reported only in smaller case series. The purpose of this study is to analyze in the largest melanoma lymphadenectomy cohort to date the risk factors that have significant association with development of post-operative complications outside of lymphedema or lymphocele. 

Methods:   This is a retrospective review of a prospectively collected database cohort of consecutive patients having either an axillary lymph node dissection (ALND) or inguinal lymph node dissection (ILND) at our institution since January 2005. Complications were defined as occurring in the first 30 days after surgery and not including lymphocele or lymphedema. Exclusionary criteria for this study include patients with bilateral dissections or those having both an ILND and an ALND. Patients were stratified according to presence or absence of complications associated with their surgery and analyzed for associated risk factors using Chi Squared tests, and Student T tests with significance defined as p< 0.05. 

Results:  Of the 524 patients who were included in our study: 175 patients (33.4%) had post-op complications. Complications were more likely to occur in females compared to males (38.5% vs 30.1%; p=0.046), those having a right-sided procedure compared to the left (38.1% vs 28.3%; p=0.017), and those having an ALND vs. ILND (50.4% vs 20.9%; p= <0.001). Other risk factors significantly correlating with post-op complications included pre-operative hypertension (39.6% with vs 29.0% without; p=0.011), or patients with a readmission within 30 days after surgery (84.2% of those readmitted had a post-op complication as the reason for re-admission vs 24.8% readmitted without a post-op complication; p=<0.001). Interestingly having a micrometastasis in the sentinel lymph node biopsy (vs. macroscopic disease and having no positive nodes on the dissection each decreased the chance of developing a post-op complication (p=0.003). Patients who developed lymphedema or a lymphocele after their surgery were also more likely to develop another post-operative complication (p=0.002 and <0.001 respectively).  Finally the number of lymph nodes removed did not correlate with any significant increased risk of developing post-operative complications. 

Conclusion:   This is the largest series to date evaluating risk factors for post-operative complications following ALND and ILND for melanoma.  Risk factors correlating with higher post-op complication rates included being female, right sided operations, pre-op HTN, or development of lymphedema or lymphocele. Protective factors included microscopic disease in the sentinel node or lack of additional positive nodes in the lymphadenectomy specimen. These factors are important and should be included in pre-operative discussions with patients having ALND and ILND for metastatic melanoma. 

 

44.01 Greater Lymph Node Retrieval and Lymph Node Ratio Impacts Survival in Resected Pancreatic Cancer

Posted on January 26, 2017

K. A. Mirkin1,2, C. S. Hollenbeak1,2, J. Wong2  1Penn State University College Of Medicine,Department Of Public Health Sciences,Hershey, PA, USA 2Penn State University College Of Medicine,Department Of Surgery,Hershey, PA, USA

Introduction:
Surgical resection is the mainstay of pancreatic cancer treatment, however, the ideal lymphadenectomy remains unsettled.  Several meta-analyses have concluded that extended lymphadenectomies do not impact survival.  The objective of this study was to determine if number of examined lymph nodes (eLN), a proxy for lymphadenectomy, and lymph node ratio (LNR) impact survival.

Methods:

The U.S. National Cancer Data Base (2003-2011) was reviewed for patients with clinical stage I and II resected pancreatic adenocarcinoma.  Patients who received neoadjuvant therapy were excluded.  Patients were stratified by eLN: 0-6, 7-12, 13-15, and >15, and LNR (LNR= #positive nodes/ #eLN):0, 0-0.2, 0.2-0.4, 0.4-0.8, and >0.8.  Univariate and multivariate survival analyses were performed.

 

Results:

14,807 patients with clinical stages I-II resected pancreatic adenocarcinoma were included.  Of these, 15.6% (N=2,309) of patients had 0-6 eLN, 27.1% (N=4,012) had 7-12, 13.4% (N=1,977) had 13-15, and 38.6% (N=5,709) had >15 eLN.  The majority of patients underwent pancreaticoduodenectomy (N=7,720, 55.1%), while the remainder underwent distal pancreatectomy (N=1,833, 13.1%), total pancreatectomy (N=1,850, 13.2%) and other (N=2,604, 18.6%).  Patients who underwent pancreaticoduodenectomy had a median eLN of 11, while those who underwent distal or total pancreatectomy or another procedure had 14. Patients with >15 eLN had significantly improved survival over the other cohorts in both node negative and node positive disease (P<0.001, both).   After controlling for patient, disease, and treatment characteristics, patients with 7-12, 13-15, and >15 eLN had improved survival relative to patients with 0-6 eLN (HR 0.87, p<0.001, HR 0.89, p=0.002, HR 0.82, p<0.001, respectively).

32.6% (N=4,829) of patients had a LNR of 0, 29.8% (N=4,414) £0.2, 9.2% (N=2,843) 0.2-0.4, 11.0% (N=1,643) 0.4-0.8, and 1.9% (N=278) had a LNR >0.8.  Patients with LNR 0 had improved survival in T1-T3 disease (P<0.01).  After controlling for patient, disease and treatment characteristics, higher LNR was negatively associated with survival (LNR 0-0.2: HR 1.44, p<0.001, LNR 0.2-0.4: HR 1.82, p<0.001, LNR 0.4-0.8: 2.03, p<0.001, LNR >0.8, p<0.001).

Even when a suboptimal number of lymph nodes were examined (eLN £6 or £12), higher LNR remained an independent predictor for mortality.

Conclusion:

Greater lymph node retrieval in stage I & II pancreatic adenocarcinoma appears to have therapeutic and prognostic value, even in node-negative disease, suggesting a comprehensive lymphadenectomy is beneficial.  Lymph node ratio is inversely related to survival and may be useful when suboptimal lymph node retrieval is performed.

43.20 Functional Recovery in Transfemoral Versus Transapical Transcatheter Aortic Valve Replacement

Posted on January 26, 2017

N. K. Asthana1, A. Mantha4, G. Vorobiof3, P. Benharash2  1University Of California – Los Angeles,Los Angeles, CA, USA 2University Of California – Los Angeles,Cardiothoracic Surgery,Los Angeles, CA, USA 3University Of California – Los Angeles,Cardiology,Los Angeles, CA, USA 4University Of California – Irvine,Orange, CA, USA

Introduction: Transcatheter aortic valve replacement (TAVR) has greatly improved treatment options for severe aortic valve stenosis patients (AS) at high surgical risk. Typically, a transfemoral (TF) approach is preferred due to being less invasive than a transapical (TA) approach. However, in patients where peripheral access is limited due to tortuosity, size, or calcification, a TA approach is preferred. This study assessed whether myocardial functional recovery differed significantly post-TAVR between patients who received a TF approach vs. a TA approach.

Methods: Echocardiograms of all severe AS patients that underwent TAVR at Ronald Reagan UCLA Medical Center from 2012-2016 were evaluated. Parameters that were assessed include left ventricular ejection fraction (LVEF), left ventricular internal diameter (LVID), interventricular septal thickness at end-diastole (IVSd), and posterior wall thickness at end-diastole (PWd). Moreover, left ventricular segmental longitudinal strains and global longitudinal strain (GLS) were measured using two-dimensional speckle tracking echocardiography (2D-STE). Echocardiograms were evaluated pre-TAVR (mean: 20.1 d), post-operatively (mean: 2.5 d), and at a 1-month follow-up (mean: 32.7 d). Statistical analysis was conducted using a repeated measures analysis of variance (rANOVA), where p < .05 was considered significant.

Results: Of the 216 patients assessed, 42 patients had complete data available. Patients that underwent TAVR with a TF approach (N = 31, 67% male, 81.6 y in age) were compared to those that underwent a TA approach (N = 11, 55% male, 87.3 y in age). For the entire cohort, between the pre-TAVR baseline and the 1-month follow-up: (i) There were no significant changes in LVEF, LVID, IVSd, or PWd (p > .05). (ii) Segmental longitudinal strains significantly increased in the apex (from -18.9 to -21.5%, p < .0001), anterior segments (from -15.5 to 18.3%, p < .0001), lateral segments (from -14.0 to -17.1%, p < .0001), inferior segments (from -14.9 to -18.1%, p < .0003), and septal segments (from -14.2 to -16.9%, p < .0002). (iii) GLS significantly improved (from -15.6 to ?18.2%, p < .001). When comparing the TF and TA groups, there were no significant differences in LVEF, LVID, IVSd, PWd, GLS, and anterior, lateral, inferior, and septal segmental longitudinal strains (p > .05). However, there was a significant difference in longitudinal strain at the apex between the TF and TA groups (TF vs. TA at 1-month follow-up: -22.3 +/- 7.63% vs. -15.9 +/- 7.47% respectively, p < .05).

Conclusion: Patients that underwent a TF approach showed significantly greater post-TAVR improvement in apical longitudinal strain, although myocardial functional recovery did not significantly differ between TF vs. TA groups otherwise. Additionally, it appears that myocardial strains measured by 2D-STE are more sensitive in detecting subclinical functional changes compared to more customary measures of cardiac remodeling.

 

43.19 Patterns of Mediastinal Metastasis after Robotic-Assisted Lobectomy for Non-Small Cell Lung Cancer

Posted on January 26, 2017

R. Gerard4, F. O. Velez-Cubian2, E. P. Ng4, C. C. Moodie1, J. R. Garrett1, J. P. Fontaine1,2,3, E. M. Toloza1,2,3  1Moffitt Cancer Center,Thoracic Oncology,Tampa, FL, USA 2University Of South Florida Morsani College Of Medicine,Surgery,Tampa, FL, USA 3University Of South Florida Morsani College Of Medicine,Oncologic Sciences,Tampa, FL, USA 4University Of South Florida,Morsani College Of Medicine,Tampa, FL, USA

Introduction:   Many thoracic surgeons perform mediastinal lymph node (LN) sampling (MLNS) in order to minimize morbidity believed to be associated with complete mediastinal LN dissection (MLND).  In order to focus attention of MLNS to the most likely LN levels involved for a given lung cancer, we sought to determine the patterns of mediastinal LN metastasis found after robotic-assisted video-thoracoscopic pulmonary lobectomy for non-small cell lung cancer (NSCLC).

Methods:   We retrospectively analyzed prospectively collected data for all patients who underwent robotic-assisted pulmonary lobectomy for NSCLC by one surgeon over 69 months.  Clinical stage was determined by history & physical examination, computerized tomography, positron-emission tomography, brain imaging studies, and/or endobronchial ultrasonography. Pathologic stage was based on intraoperative findings and final pathology. The pulmonary lobe resected and any mediastinal LNs involved by metastasis were noted.

Results:  Of 303 NSCLC patients (pts), mean age was 69±0.5 yr (range 39-98 yr), with most common histologies being adenocarcinoma (66%), squamous cell carcinoma (21%), and neuroendocrine carcinoma (10%).  Tumors were located in the right lung in 198 (65.3%) pts and in the left lung in 105 (34.7%) pts.  The three most common anatomic locations were right upper lobe (RUL; 39.6%), left upper lobe (LUL; 21.8%), and right lower lobe (RLL; 18.5%).  Frequencies of stage-3 disease were similar for left NSCLC compared to right NSCLC (p=0.59), but the frequency of stage-2 disease was higher for left NSCLC (28.6%) compared to that for right NSCLC (17.2%; p=0.02).  Of stage-3A right NSCLC, 56.8% were in the RUL, while 69.6% of stage-3A left NSCLC were in the LUL.  Among N1 LNs, level 11 involvement was more common than level 10 involvement for all right and left NSCLC combined (72/103, 69.9% vs. 20/103, 19.6%; p<0.0001).  Mediastinal LN involvement was highest in level 4R (23/198; 11.6%), level 5 (11/105; 10.5%), level 7 (25/303; 8.3%), and level 2R (10/198; 5.1%).  Stage-3A RLL NSCLC most commonly metastasized to level 7 (12/26; 46.2%), while stage-3A left lower lobe NSCLC metastasized most commonly to level 9L (3/6; 50.0%).

Conclusion:  After robotic-assisted pulmonary lobectomy, mediastinal LN metastatic disease was similarly frequent for right versus left NSCLC, while stage-2 disease was more frequent with left NSCLC.  Among N1 LNs, interlobar LNs were more commonly involved than hilar LNs.  For stage-3A NSCLC, there was upper lobe predominance on both sides.  Level 4R LNs were the most frequently found to be positive with right NSCLC, mostly due to RUL NSCLC, while level 5 LNs were most frequently found to be positive with left NSCLC, mostly due to LUL NSCLC.  These patterns of N1 and mediastinal LN involvement should assist in guiding thoracic surgeons to perform a more focused MLNS or a more complete MLND for more accurate NSCLC staging.

43.18 Predictors of Cardiogenic Shock in Cardiac Surgery Patients Receiving Intra-Aortic Balloon Pumps

Posted on January 26, 2017

A. Iyengar1, O. Kwon2, R. Shemin2, P. Benharash2  1University Of California – Los Angeles,David Geffen School Of Medicine,Los Angeles, CA, USA 2University Of California – Los Angeles,Division Of Cardiac Surgery,Los Angeles, CA, USA

Introduction:  Cardiogenic shock following cardiac surgery is a rare complication that leads to increased morbidity and mortality. Intra-aortic balloon pumps (IABPs) may be used during the perioperative period to increase coronary perfusion and support cardiac output. The purpose of this study was to characterize predictors of postoperative cardiogenic shock in cardiac surgery patients, and examine differences between those with and without IABP support.

Methods:  Retrospective analysis of UCLA’s Society of Thoracic Surgeon’s (STS) database was performed between January 2008 and July 2015. Preoperative demographic data for all patients were queried, and patient’s receiving IABP support during the perioperative period were identified. The Kruskal-Wallis and chi-squared tests were used for comparisons between IABP and control cohorts. Multivariable logistic regression with step-wise elimination was used to model postoperative cardiogenic shock in both the IABP and control cohorts. 

Results: During the study period, 4,741 cardiac surgery patients were identified during the study period, of whom 268 (6%) received an IABP. IABP patients had higher rates of previous cardiac surgery (54% vs. 38%, p<0.001), congestive heart failure (69% vs. 43%, p<0.001), and preoperative cardiogenic shock (22% vs. 2%, p<0.001). Furthermore, IABP patients were more likely to have emergent operations (84% vs. 42%, p<0.001) and receive coronary artery bypass grafts (CABG, 63% vs. 32%). IABP patients had significantly greater ventilation times, ICU/total hospital stays, and 30-day mortality (all p<0.001), and more postoperative cardiogenic shock (10% vs. 3%, p<0.001).
Among the IABP cohort, preoperative dialysis, arrhythmias, and previous cardiac surgery were all associated with higher odds of postoperative cardiogenic shock (all p<0.10), while CABG operations were found to be protective compared to other cardiac operations (OR 0.33 vs 2.28, p=0.008 and 0.053, respectively). On multivariate analysis, previous cardiac surgery and preoperative arrhythmia remained significant (AOR 5.95, p=0.005, and 2.94, p=0.015, respectively)predictors of postoperative cardiogenic shock. In the control cohort, several factors including hypertension, chronic lung disease, preoperative congestive heart failure, cardiogenic shock, inotropic medications, urgent/emergent status, non-CABG/Valve cardiac surgery, and prolonged bypass times, were associated with postoperative cadiogenic shock.

Conclusion: Factors associated with cardiogenic shock among post-cardiac surgery patients differ between those patients receiving IABP and those who do not. Among IABP patients, previous cardiac surgery and arrhythmias were associated with increased rates of cardiogenic shock, while shock was multifactorial among control patients. The etiology of cardiogenic shock may differ between these two cohorts, and early identification of those patients at risk may lead to improved outcomes.

43.17 Tumor Size and Perioperative Outcomes after Robotic-Assisted Pulmonary Lobectomy

Posted on January 25, 2017

R. Gerard4, F. O. Velez-Cubian2, E. P. Ng4, C. C. Moodie1, J. R. Garrett1, J. P. Fontaine1,2,3, E. M. Toloza1,2,3  1Moffitt Cancer Center,Thoracic Oncology,Tampa, FL, USA 2University Of South Florida Morsani College Of Medicine,Surgery,Tampa, FL, USA 3University Of South Florida Morsani College Of Medicine,Oncologic Sciences,Tampa, FL, USA 4University Of South Florida,Morsani College Of Medicine,Tampa, FL, USA

Introduction:   Tumor size is one factor that determines whether lobectomy is performed via open or minimally invasive approach.  We investigated whether tumor size affects perioperative outcomes after robotic-assisted video-thoracoscopic (RAVT) pulmonary lobectomy.

Methods:   We retrospectively studied all patients (pts) who underwent RAVT pulmonary lobectomy between September 2010 and May 2016 by one surgeon at our institution.  Patients were grouped by greatest tumor diameter on pathologic measurement of lobectomy specimens.  Perioperative outcomes, including estimated blood loss (EBL), skin-to-skin operative time, conversion to open lobectomy, intraoperative and postoperative complications, chest tube duration, hospital length of stay (LOS), and in-hospital mortality were compared.  Chi-square test, Student’s t-test, and Kruskal-Wallis test were used, with p≤0.05 as significant.

Results:  We identified and grouped 359 pts by greatest tumor diameter being ≤10mm, 11-20mm, 21-30mm, 31-50mm, or ≥51mm.  Tumor histology was comprised of NSCLC (89.4%), SCLC (1.9%), and pulmonary metastases (8.6%), with the most common NSCLC histology being adenocarcinoma (63.8%), squamous cell (21.5%), and neuroendocrine (9.7%).  No differences were noted in mean age, female:male ratio, or mean body surface area among the groups, but mean body mass index was lowest in pts with tumors ≥51mm.  Lobar distribution of lung tumors did not differ among the groups (p>0.14), but extent of resection differed by pts with tumors ≥51mm having a lower rate of simple lobectomies (p<0.001) and a higher rate of en bloc chest wall resection (p<0.001).  Neither overall intraoperative complications nor overall or emergent conversion to open lobectomy differed among the groups (p>0.21), but pulmonary artery (PA) injury occurred in as high as 7.2% of pts in groups with tumors ≥21mm (p=0.014).  While median EBL was higher in pts with tumors ≥51mm (p≤0.003) and median operative time was higher in pts within groups with tumors ≥31mm (p≤0.019), median chest tube duration and median hospital LOS did not differ among the groups (p>0.37).  Neither overall total postoperative complications nor overall pulmonary or cardiovascular complications differed among the groups (p>0.23), but pneumothorax after chest tube removal and requiring intervention was more frequent in pts who had tumors ≤10mm (p=0.03).  In-hospital mortality did not differ among the groups (p=0.60).

Conclusions: Patients who undergo RAVT lobectomy for tumors ≥51mm are associated with lower BMI and are less likely to have simple lobectomies and more likely to require en bloc chest wall resection.  Patients with larger tumors also are at increased risk of PA injury, higher EBL, and longer operative times, but are at lower risk for pneumothorax after chest tube removal and that require intervention.  However, tumor size does not affect chest tube duration, hospital LOS, or in-hospital mortality.

43.16 Weekend Discharge and Readmission Rates After Cardiac Surgery

Posted on January 25, 2017

G. Ramos1, R. Kashani1, Y. Juo1, A. Lin2, N. Satou1, R. J. Shemin1, P. Benharash1  1David Geffen School Of Medicine, University Of California At Los Angeles,Division Of Cardiac Surgery,Los Angeles, CA, USA 2David Geffen School Of Medicine, University Of California At Los Angeles,Division of General Surgery,Los Angeles, CA, USA

Introduction:  Unintended rehospitalization within 30 days serves as a quality metric for institutions and may lead to financial penalties. Few studies have examined the implications of weekend discharges on readmission rates. Limited care coordination and cross coverage of surgeons are known challenges of weekend hospital function and may lead to less comprehensive post-discharge care plan. Based on this knowledge, we hypothesized that patients discharged on weekends would be more likely to be readmitted. 

Methods:  Using the institutional Society of Thoracic Surgeons (STS) database, all adult patients (>18) undergoing cardiac surgery between 2008 and 2015 were identified. 44 demographic and perioperative characteristics were collected and accounted for in a multivariate model. Emergency, transplant and mechanical assist patients were excluded. Weekday discharge was defined as being discharged on Monday through Friday while weekend discharge was limited to Saturdays and Sundays. The primary outcome variable was any readmission within 30 days of discharge. 

Results: Of the 4416 patients included in the study, 3632 (82%) were discharged on a weekday, and 783 (18%) were discharged on a weekend; 495 (11%) patients were readmitted within 30 days. The readmission rates for the weekday and weekend cohorts were similar (10% vs.11%, p=0.4). After adjustment for other risk factors of readmission, there was no association between discharge day of the week and readmission (adjusted odds ratio [AOR]= 1.08, 95% CI=0.83-1.42, p=0.6).  Significant risk factors for readmission included receiving dialysis (AOR=1.6, 95% CI=1.09-2.30, p=0.016), undergoing an urgent operation (AOR=1.24, 95% CI=1.01-1.53, p=0.04), and taking preoperative Coumadin (AOR=1.43, 95% CI=1.04-1.97, p=0.03) or beta-blockers (AOR=1.32, 95% CI=1.07-1.63, p=0.01).

Conclusion: In this study, weekend discharge status was not associated with an increased risk of readmission after cardiac surgery. However, patients having an urgent surgery, receiving dialysis, or taking preoperative Coumadin or beta-blockers were more likely to be readmitted. Our findings suggest that readmission reduction programs should focus on patient factors rather than providing additional weekend coverage beyond existing care coordination resources. 

 

43.15 Impact of Prior Myocardial Infarction on Myocardial Recovery after Transcatheter Aortic Valve Replacement

Posted on January 25, 2017

A. Mantha2, N. Asthana3, G. Vorobiof3, P. Benharash1  1University Of California – Los Angeles,Cardiac Surgery,Los Angeles, CA, USA 2University Of California – Irvine,Orange, CA, USA 3University Of California – Los Angeles,Cardiology,Los Angeles, CA, USA

Introduction:  Transcatheter valve replacement (TAVR) is a definitive, minimally invasive treatment for patients with severe stenosis (AS) and has been shown to improve myocardial remodelling. However, it is unclear whether these changes occur in patients who have a history of myocardial infarction. This study sought to evaluate the impact of TAVR in patients who had previously suffered a myocardial infarction and have undergone PCI or CABG.

Methods:  A review of our prospectively-maintained institutional Society of Thoracic Surgeons database and Transcatheter valve Registry was performed to identify all patients undergoing TAVR from Jan. 2013- Mar. 2016. Chi-square test and regression were used to evaluate differences in patient demographics, readmission rate and length of stay. Repeated measures analysis of variance was used to compare myocardial strain and function among forty-two patients with speckle data through one-month follow-up. 

Results: Of the 172 patients included in the analysis, 62 (36%) had previously experienced a myocardial infarction. 22 (33%) underwent PCI alone, 18 (27%) underwent CABG alone and 9 (13%) underwent both PCI and CABG. Patients who underwent PCI alone had a significantly longer length of stay (7.2 vs 4.4 days, p<0.001) after TAVR and higher proportion of patients readmitted within 30 days (37%, p<0.01). Patients with history of MI had consistently lower magnitudes of strain in the septal (p<0.01), anterior (p<0.02), lateral (p<0.01), and inferior (p<0.01) ventricular walls despite having similar ventricular diameter (p=0.77) and septal thickness (p=0.82). Both cohorts demonstrated significant improvement in global longitudinal strain (-19.0 vs -16.0 in control, -17.1 vs -14.1 in MI, p<0.01) and interaction between history of MI and GLS was not significant.

Conclusion: Patients with history of myocardial infarction benefit similarly from TAVR as control patients with no history of MI despite having poorer pre-operative ventricular contractility. However, management strategy of infarction episode may have differential impact on tissue ischemia leading to increased length of stay and risk of readmission after TAVR.
 

43.14 Thoracoscopic Lobectomy Reliability for NSCLC is an Important Indicator of Program Development

Posted on January 25, 2017

M. Hennon1,2, J. Xiao2, M. Huang1, E. Dexter1,2, A. Picone1,2, S. Yendamuri1,2, C. Nwogu1,2, W. Tan3, T. Demmy5  1Roswell Park Cancer Institute,Thoracic Surgery,Buffalo, NY, USA 2State University Of New York At Buffalo,Surgery,Buffalo, NY, USA 3Roswell Park Cancer Institute,Biostatistics,Buffalo, NY, USA 5Rutgers Cancer Institute Of New Jersey,Cardiothoracic Surgery,New Brunswick, NJ, USA

Introduction:  Outcomes for thoracoscopic (VATS) lobectomy at the institutional level can be affected by numerous variables, including selection bias. The total percentage of cases completed by VATS for locally advanced nonsmall cell lung carcinoma may be an important component of individual program quality.

Methods:  Over 11 years (from January 2002 to March 2013), 1289 consecutive lobectomies were performed, of which 300 were for patients with locally advanced NSCLC (tumors greater than 4cm, T3, T4, or patients who underwent induction chemotherapy).  Patients requiring chest wall resection, sleeve lobectomy, or pneumonectomy were excluded.  Cases were divided into three sequential groups of 100 patients for comparison.  Reliability is defined as the total number of cases completed thoracoscopically (VATS) divided by all cases (VATS + Conversion+ Open). Conversion rates, percentage of cases completed by VATS, along with preoperative, perioperative and outcome variables were compared and analyzed by Mann-Whitney-Wilcoxon and Fisher’s exact tests. Estimated overall survival and disease free survival distributions were obtained using the Kaplan-Meier method.

Results: Of 300 cases during the study period, 219 were completed by VATS.  VATS reliability increased from 62% (early), to 77% (middle), and 80% (late).  Reliability increased due to a steady decrease in planned thoracotomy from 17%, to 9% and 2.1% respectively.  A higher percentage of patients in the late group had more preoperative comorbidities (CAD/MI 27% vs. 19% vs. 42.6%, p = 0.0016). Median operative time increased over the study period from 225 min. [96-574] vs. 328 min. [115-687] vs. 340 min. [140-810], presumably due to approaching more complex tumor pathology.  Median operative blood loss was the same for all groups at 200 mL (10-2200).  Median postoperative ICU stay was 1 day (0-92) for all groups.  Higher neoadjuvant therapy rates (16% vs. 54% vs. 50%, p <0.0001) were achieved in the middle/late groups.  Fewer postoperative complications occurred in the middle and late time groups (any major complication was 38% vs. 13% vs. 16%, p < 0.0001; bleeding 23% vs. 4% vs. 6%, p < 0.0001; air leak 16% vs. 13% vs. 3%, p = 0.0037).   Number of lymph nodes harvested during surgery (10.2 vs. 12.5 vs. 22.8, p <0.0001) improved significantly. 

Conclusion: In our experience, VATS reliability increased over time with favorable perioperative and postoperative outcomes due to fewer cases being approached by planned thoracotomy. Since there were associations with factors like lymph node harvest, VATS reliability deserves additional study as an indicator of individual program achievement and as a tool to explain differences between VATS and open surgeries reported in large, cooperative databases.

 

43.13 The Use of Peri-operative Ketorolac in Surgical Treatment of Pediatric Spontaneous Pneumothorax

Posted on January 25, 2017

R. M. Dorman1,2, G. Ventro1,2, S. Cairo1, K. Vali1,2, D. H. Rothstein1,2  1Women And Children’s Hospital Of Buffalo,Pediatric Surgery,Buffalo, NY, USA 2State University Of New York At Buffalo,Pediatric Surgery,Buffalo, NY, USA

Introduction:
The effect of post-opertative anti-inflammatory medications on pleurodesis success after treatment of spontaneous pneumothorax is uncertain. We sought to determine if the use of post-operative ketorolac is associated with an increased risk of recurrence in the surgical treatment of primary spontaneous pneumothorax in children.

Methods:
The Pediatric Health Information System database was queried for all patients age 10-16 years discharged ibetween 2004-2014 with a primary diagnosis of pneumothorax or pleural bleb and a thoracotomy, thoracoscopy, or lung resection procedure. Deaths, encounters representing readmission after previous operative treatment of pneumothorax in the prior year, patients requiring extra-corporeal life support, and patients with diagnoses or concurrent procedures that may lead to secondary or iatrogenic pneumothorax were excluded. Variables included basic demographics, discharge in the first or second half of the study period, chronic renal or hematologic disease, intensive care unit admission or post-operative mechanical ventilation, and whether a lung resection or plication was coded. The primary predictor of interest was ketorolac administration any time in the period from post-operative day 0 to 5. The primary outcomes of interest were thoracentesis, thoracostomy, thoracotomy, thoracoscopy, lung resection or plication, or pleurodesis within 1 year of the index admission. Bivariate analyses were carried out for all outcomes and multivariate logistic regression analyses were then performed for reintervention and readmission.  

Results:
1,678 records met inclusion criteria. Three hundred ninety-five (23%) were subsequently excluded (227 readmissions and the remainder for one of the above-listed criteria), leaving 1,283 patients for analysis. The cohort was predominately male (79%), white (74%), and older (mean age 15.5 ± 1.2 years). Most patients had some lung resection recorded (78%), a majority were administered ketorolac (57%), and few required reintervention (20%) or readmission (18%). Mean postoperative length of stay was 5.2 ± 3.8 days and mean cost was $17,649 ± $10,599. Older patients and those in the earlier years of the study were more likely to receive ketorolac. There was significant variation in frequency of ketorolac administration by geographic region, ranging from 32% to 68%. On multivariate analysis, no variable was predictive of reintervention, and only lung resection correlated with readmission (adjusted odds ratio 0.63 [95% C.I. 0.45-0.90]). 

Conclusion:
Post-operative ketorolac administration was not associated with an increased likelihood of reintervention or readmission within 1 year of operative treatment of primary spontaneous pneumothorax, suggesting that it may be used safely as part of a post-operative pain control regimen. Effects on postoperative length of stay and cost, however, were not demonstrated.
 

43.12 Acute Retrograde Type A Aortic Dissection: Morphological Analysis and Clinical Implications

Posted on January 25, 2017

B. L. Rademacher1, P. D. DiMusto2, J. L. Philip1, C. B. Goodavish3, N. C. De Oliveira3, P. C. Tang3  1University Of Wiscosin,Department Of Surgery, Division Of General Surgery,Madison, WISCONSIN, USA 2University Of Wisconsin,Department Of Surgery, Division Of Vascular Surgery,Madison, WISCONSIN, USA 3University Of Wiscosin,Department Of Surgery, Division Of Cardiothoracic Surgery,Madison, WISCONSIN, USA

Introduction: Numerous studies have described thoracic stent graft induced retrograde type A dissections (rTAD), however, much less is known about acute spontaneous rTAD with tears originating past the left subclavian without prior aortic instrumentation. This study compares the morphology of acute rTAD with both acute antegrade type A dissection (aTAD) with primary tears in the ascending aorta and acute type B dissection.

Methods: From 2000 to 2016, there were 12 acute rTAD, 96 aTAD, and 92 acute type B dissections with available imaging that underwent operative intervention at our institution. Dissection morphology along the length of the aorta was characterized using 3-dimensional reconstruction based on computerized tomography angiography (CTA) images. We examined primary and secondary tear characteristics, true lumen area as a fraction of the total lumen area, and false lumen contrast intensity as a fraction of the true lumen contrast intensity.  Features of presentation and operative parameters were compared between rTAD and aTAD.

Results: Compared with acute type B dissections, primary rTAD tears were more common in the distal arch (75% vs 43%, p=0.04), and the false lumen to true lumen contrast intensity ratio at the mid-descending thoracic aortic level was lower (0.46 vs 0.71, P=0.02) indicating more sluggish blood flow or thrombosis in the false lumen. rTAD cases had less decompression of the false lumen compared with acute type B dissections such that there were fewer aortic branch vessels distal to the subclavian that were either exclusively perfused through the false lumen or through both the false and true lumen (0.40 vs 2.19, P<0.001). Compared with aTAD, rTAD had a tendency for less root involvement where true lumen as a fraction of total lumen area at the root level was higher (0.88 vs 0.76, P=0.081). rTAD had a lower false lumen to true lumen contrast intensity ratio compared to aTAD at the root (0.25 vs 0.57, P<0.05), ascending aorta (0.25 vs 0.72, P<0.001), and proximal arch (0.39 vs 0.67, P<0.05) indicating more sluggish flow or greater tendency to thrombose. rTAD patients were more likely to undergo aortic valve resuspension (100% vs 74%, P=0.044) than aortic valve replacement, and tend to have lower aortic cross-clamp times (83 vs 108 min, P=0.066) (Table 1).

Conclusion: This study suggests that retrograde propagation of the false lumen to the arch and ascending aorta tends to occur when the primary tears that occur distal to the left subclavian are in close proximity to the aortic arch and when false lumen decompression through the distal aortic branches are less effective. Compared to aTAD, rTAD tends to have less root involvement and successful aortic valve resuspension is more likely.

« Previous Page
Next Page »