Alpha-1 Antitrypsin Augmentation Therapy in Chronic Pancreatitis Patients Undergoing Total Pancreatectomy and Islet Autotransplantation: A Randomized, Controlled Study.

Stress-induced islet graft loss during the peri-transplantation period reduces the efficacy of islet transplantation. In this prospective, randomized, double-blind clinical trial, we evaluated the safety and efficacy of 60 mg/kg human alpha-1 antitrypsin (AAT) or placebo infusion weekly for four doses beginning before surgery in chronic pancreatitis (CP) patients undergoing total pancreatectomy and islet autotransplantation (TP-IAT). Subjects were followed for 12 months post-TP-IAT. The dose of AAT was safe, as there was no difference in the types and severity of adverse events in participants from both groups. There were some biochemical signals of treatment effect with a higher oxygen consumption rate in AAT islets before transplantation and a lower serum C-peptide (an indicator of islet death) in the AAT group at 15 min after islet infusion. Findings per the statistical analysis plan using a modified intention to treat analysis showed no difference in the C-peptide area under the curve (AUC) following a mixed meal tolerance test at 12 months post-TP-IAT. There was no difference in the secondary and exploratory outcomes. Although AAT therapy did not show improvement in C-peptide AUC in this study, AAT therapy is safe in CP patients and there are experiences gained on optimal clinical trial design in this challenging disease.

Neoadjuvant Chemotherapy is Associated With Decreased Survival in Early-Stage Gastric Cancer.

BACKGROUND: Although randomized controlled trials on neoadjuvant chemotherapy for gastric cancer have included some T1-staged tumors, overall survival (OS) has not been analyzed for this subset. Due to the low negative predictive value of clinical staging and the benefits of neoadjuvant chemotherapy for locally advanced disease, identifying patient groups with early-stage gastric cancer that may benefit from neoadjuvant chemotherapy is of merit. AIMS: The objective of this study was to evaluate the relationship between OS and sequence of surgical therapy for clinical T1 gastric cancer. METHODS: The 2017 National Cancer Database was used to compare patients who had surgery-first and those who received neoadjuvant chemotherapy for T1-stage gastric cancer. OS was analyzed using a parametric regression survival-time model adjusted for covariates. The effects of these covariates on OS based on surgical sequence were examined. RESULTS: 11,219 patients were included, of which 10,191 underwent surgery as their first or only treatment. When adjusted for covariates, neoadjuvant chemotherapy followed by curative-intent surgery was significantly associated with increased risk of death (HR 1.15, 95% CI 1.01-1.31, P = .030). In multivariate analysis, clinical N0 stage, non-minorities, and patients with high socioeconomic status had improved OS if they did not have neoadjuvant chemotherapy and instead had upfront surgery. CONCLUSION: Neoadjuvant chemotherapy is associated with decreased OS for early-stage gastric adenocarcinoma, even for patients with clinically positive nodal disease. In addition, the lack of survival improvement with a surgery-first approach in patients with disparities deserves further study.

Overall Survival Differences in Young Black Colorectal Cancer Patients: a Report from the National Cancer Database.

Objectives: Black patients have the highest overall incidence rate of early onset colorectal cancer, with many of these patients presenting with more aggressive disease at diagnosis, ultimately leading to decreased overall survival. We aimed to (1) evaluate how race and age affected overall survival in colorectal cancer patients, and (2) determine the different demographic and clinical covariables that may influence survival in younger individuals. Methods: The 2017 National Cancer Database (NCDB) was used to identify all patients that had colorectal cancer between 2004-2017. These patients were then divided into groups according to age (<45 and ≥45 years old) and race (white and black). Overall survival (OS) between white and black groups according to age was compared. Initial testing of survivor functions between groups revealed violations of the proportional hazards assumption. Accordingly, we used parametric maximum likelihood analyses fitting the survivor functions to Weibull distributions. Logistic regression analysis was used to determine univariate and multivariate relationships between the covariates and race for younger subjects. Propensity score matching analysis was also used to control for differences in the demographic or clinical variables between the young black versus white subgroups. Results: Out of 1.4 million potential cases initially identified, 207,823 unique cases were deemed eligible for evaluation based on study criteria. Black patients in the study population were more likely to be female, have medical comorbidities, and come from areas with lower average income and baseline education. OS was lower in older patients of both race categories when compared to the younger cohorts. Among patients older than 45 years, there were no significant differences in proportional hazard of death between black and white patients. However, among those younger than 45 years, younger black patients had significantly increased hazard of death. Regarding disease burden at diagnosis, pathologic characteristics and overall risk of death, there were no significant differences between black and white patients. Conclusions: Overall survival in young black patients with colorectal cancer is significantly reduced when compared to young white patients, even when controlling for demographic and pathologic factors. This suggests that the outcome disparities between black and white patients are complex, and the underlying factors are not well understood.

An Early-Stage Colon Cancer Develops Intrabiliary Growth Type Metastasis 7 Years After Curative Colectomy.

Although liver metastasis commonly occurs in patients with colorectal cancer (CRC), it is infrequent that it presents several years after curative resection for early-stage disease. Even more unusual is development of intrabiliary growth type metastasis rather than parenchymal metastasis. When this occurs, it can be mistaken for cholangiocarcinoma. We present a case in a patient with history of pT1N0M0 CRC treated with sigmoidectomy 7 years previously who presented with abdominal pain and MRI revealing left hepatic ductal dilation with no accompanying mass. With a recent normal colonoscopy and carcinoembryonic antigen, he was diagnosed with cholangiocarcinoma. Anatomic hepatic resection was performed, and final pathology with immunohistochemistry revealed staining consistent with CRC metastasis rather than cholangiocarcinoma. Intrabiliary growth type metastasis is a rare occurrence, which leads to its misdiagnosis. Patients with an intrabiliary tumor and a history of CRC should have immunohistochemistry to confirm the diagnosis to ensure appropriate adjuvant treatment and counseling.

The Effect of Goal-Directed Fluid Administration on Outcomes After Pancreatic Surgery.

BACKGROUND: We evaluated the effect of an Enhanced Recovery After Surgery protocol on intraoperative fluid administration and postoperative outcomes in pancreatic surgery. METHODS: Pancreatic cancer resections at our institution from 2012 to 2018 were grouped according to pre- or post-protocol initiation. Preoperative characteristics and postoperative outcomes were compared with Fisher's exact test and chi-square for categorical variables, and Mann-Whitney U test for continuous variables. Further analysis separated patients that had a Whipple from those who had distal pancreatectomy. RESULTS: A total of 263 patients underwent pancreatic cancer resection during the study period (169 Whipples, 84 DPs, 92 pre-ERAS and 171 post-ERAS). Intraoperative fluid administration significantly decreased after protocol implementation (mean 6,277 ml vs. 3870 ml, p < 0.001). This held true when separating patients that had a Whipple procedure from those that had a DP (6,929 ml vs. 4,513 ml, p < 0.001, 5,060 ml vs. 2,833 cc, p = 0.002, respectively). Intensive care unit (ICU) admission (41.3% vs. 20.5%, p < 0.001) and length of stay (9.4 vs. 8.1 days, p < 0.01) were significantly reduced after ERAS implementation for all patients and in Whipple patients alone (47.5% vs. 23.6%, p = 0.002 and 10.7 vs. 6.6 days, p = 0.004). DP patients also had significantly decreased ICU admissions (41.3% vs. 20.5%, p = 0.045). All other postoperative outcomes were not significantly different. CONCLUSION: For patients undergoing pancreatic cancer resection, goal-directed fluid management is associated with decreased intraoperative fluid administration, decreased ICU admission, and decreased length of stay without an increase in postoperative complications or readmission.

Distal Pancreatectomy for Pancreatitis in the Modern Era.

INTRODUCTION: Distal pancreatectomy has not been well examined in the modern era to guide management for pancreatitis. We evaluated this heterogeneous group and the preoperative factors associated with clinically relevant postoperative pancreatic fistula (CR-POPF). METHODS: Patients undergoing distal pancreatectomy at a single academic institution from August 2012 to January 2020 were evaluated. Univariate and multivariate logistic regressions were conducted between preoperative factors and CR-POPF. RESULTS: One hundred and thirty patients underwent distal pancreatectomy. Indication for operative management included chronic pancreatitis and/or pseudotumor in 24.6% (n = 32), disconnected left pancreatic remnant in 31.5% (n = 41), chronic distal pseudocyst in 20.8% (n = 27), and distal necrosis in 13.8% (n = 18). Significant complications (Clavien-Dindo grade ≥ III) were seen in 34% of patients. After surgery, 34.2% developed diabetes, 40% had persistent opioid use, and 22.3% had CR-POPF. In multivariate analysis, male sex was significantly associated with CR-POPF (odds ratio 3.1, P = 0.037), and having a preoperative, therapeutic endoscopic retrograde cholangiopancreatography was protective (odds ratio 0.28, P = 0.020). CONCLUSIONS: Distal pancreatectomy is undertaken in pancreatitis with high morbidity. Female sex and preoperative, therapeutic endoscopic retrograde cholangiopancreatography were significant protective factors for CR-POPF. The natural history of this approach is relevant for those with distal pancreatitis failing medical management.

A Mouse Model for Chronic Pancreatitis via Bile Duct TNBS Infusion.

Chronic pancreatitis (CP) is a complex disease involving pancreatic inflammation and fibrosis, glandular atrophy, abdominal pain and other symptoms. Several rodent models have been developed to study CP, of which the bile duct 2,4,6 -trinitrobenzene sulfonic acid (TNBS) infusion model replicates the features of neuropathic pain seen in CP. However, bile duct drug infusion in mice is technically challenging. This protocol demonstrates the procedure of bile duct TNBS infusion for generation of a CP mouse model. TNBS was infused into the pancreas through the ampulla of Vater in the duodenum. This protocol optimized drug volume, surgical techniques, and drug handling during the procedure. TNBS-treated mice showed features of CP as reflected by bodyweight and pancreas weight reductions, changes in pain-associated behaviors, and abnormal pancreatic morphology. With these improvements, mortality associated with TNBS injection was minimal. This procedure is not only critical in generating pancreatic disease models but is also useful in local pancreatic drug delivery.

Travel distance and overall survival in hepatocellular cancer care.

BACKGROUND: Our objective was to assess the relationship between overall survival (OS) and distance travelled to the treating facility for patients undergoing liver resection for hepatocellular carcinoma and to determine whether this relationship was dependent upon the structural factors of the treating facility. METHODS: Using National Cancer Database, we focused on extremes of travel: Local (<12.5 miles to treating facility) and Travel (≥50 miles). We analyzed OS with Cox models; we estimated stratified models to assess interaction between distance and facility characteristics (volume, academic status). RESULTS: We included 6860 patients. After correction for confounding, distance travelled was not associated with OS (p = 0.444). However, Travel patients treated at high-volume, academic centers had worse OS compared to Local patients (HR 1.54, 95%CI 1.07-2.21); this association was not seen for patients treated at low volume, academic centers (p = 0.708) high volume non-academic centers (p = 0.174) or low volume non-academic centers (p = 515). CONCLUSION: For those patients treated at high-volume, academic centers, living far from the facility was associated with worse OS. The reasons for this association should be investigated further.

Travel distance and its interaction with patient and hospital factors in pancreas cancer care.

BACKGROUND: Although volume-outcome literature supports regionalization for complex procedures, travel may be burdensome. We assessed the relationship between overall survival and travel distance for patients undergoing pancreatic resection for adenocarcinoma. METHODS: We analyzed the Fall 2018 National Cancer Database Public Use File. We defined distance traveled as a categorical variable (<12.5 miles, 12.5-50mi, and >50mi). We analyzed overall survival (OS) as a function of distance traveled using the log rank test and Cox proportional hazards models; we estimated stratified models to assess for interaction between distance and other relevant covariates. RESULTS: In adjusted analysis of 39,089 patients, greater distance was associated with decreased OS (p = 0.0029). We found interactions between distance and center type, comorbidities, and age. Distance traveled was a negative factor for patients treated at low-volume academic centers (but not high-volume academic or non-academic centers). Additionally, distance traveled was a negative factor for OS in young, healthy patients but not geriatric, ill patients. CONCLUSION: Traveling more than 12.5 miles for pancreatic resection was associated with worse OS. Prior to regionalization, evaluation of local resources may be necessary.

High Mitotic Rate Predicts Sentinel Lymph Node Involvement in Thin Melanomas.

BACKGROUND: Indications for sentinel lymph node (SLN) biopsy in the population with thin melanoma have frequently changed over time. The objective of our study was to evaluate T1 melanoma pathologic features predictive of SLN positivity with a primary focus on identifying a specific mitotic value that is most predictive of lymph node disease. Further detailed predictive features would help physicians select patients with thin melanoma for SLN biopsy. METHODS: The Surveillance, Epidemiology, and End Results database was queried for all patients diagnosed with trunk or extremity cutaneous melanoma with ≤1 mm depth who underwent SLN biopsy between the years of 2010 and 2013. Patient demographics and tumor characteristics including depth, mitotic rate (MR), ulceration, and tumor location were evaluated. MR was dichotomized at multiple cut points to identify the ideal number of mitosis for MR as a predictor of SLN status. Multivariable logistic regression analyses were performed to identify the factors affecting nodal positivity and the impact of MR threshold. Kaplan-Meir curves were used for overall survival (OS) analysis. RESULTS: Factors significantly associated with SLN positivity in the entire cohort included MR (P < 0.001, OR 1.24, 95% CI 1.18-1.31), tumor location (P = 0.017, OR 1.48, 95% CI 1.07-2.05), and ulceration (P < 0.001, OR 2.01, 95% CI 1.39-2.93,). An MR ≥ 4 was significant for SLN positivity (P = 0.049, OR 1.08, 95% CI 1.01-1.38). Mean OS was 46.7 mo for MR < 4 compared with 43.2 mo for MR ≥ 4 (P < 0.001). CONCLUSIONS: MR ≥ 4 was significant and associated with SLN positivity in thin melanomas and asulceration. Thus, MR ≥ 4 should be considered as an indication for SLN biopsy in thin melanoma.

Sphincter of Oddi Dysfunction After Gastric Bypass: Surgical or Endoscopic Therapy?

BACKGROUND: Management of Sphincter of Oddi Dysfunction (SOD) requires advanced techniques (endoscopic retrograde cholangiopancreatography via gastrostomy [GERCP]) after Roux-en-Y gastric bypass (RYGB) for obesity. Transduodenal sphincteroplasty (TS) is also performed yet carries the risks of surgery. We hypothesized that TS would have increased morbidity and mortality but provide a more durable remission of symptoms. METHODS: All patients between 2005 and 2016 with RYGB for obesity undergoing endoscopic or surgical management for type I or II SOD were included in the study. Patients with type III SOD, or who underwent RYGB for nonobesity indications, were excluded. RESULTS: Thirty-eight patients were identified. GERCP was initially performed in 17 patients, whereas TS was performed in 21. Thirty-day mortality was 0% in our cohort, and 30-d morbidity was similar between GERCP and TS (29% versus 10%; P = 0.207). Resolution of symptoms after initial therapy was seen in 41% of GERCP (7/17) and 67% of TS (14/21) (P = 0.190), respectively, and overall after 35% (8/23) and 64% (16/24) of procedures performed (P = 0.042). Symptom resolution, as defined by the median ratio of days of total remission by total days of observed follow-up, was shorter after initial and all interventions with GERCP compared with TS (0.67 versus 1.00, P = 0.036 and 0.52 versus 1.00, P = 0.028, respectively). CONCLUSIONS: Endoscopic and surgical treatment of SOD had similar morbidity and mortality. However, treatment success and duration of remission was higher in those treated with surgery.

Are BRCA1 and BRCA2 gene mutation patients underscreened for pancreatic adenocarcinoma?

BACKGROUND: Breast cancer (BRCA) mutations account for the highest proportion of hereditary causes of pancreatic ductal adenocarcinoma (PDAC). Screening is currently recommended only for patients with one first-degree relative or two family members with PDAC. We hypothesized that screening all BRCA1/2 patients would identify a higher rate of pancreatic abnormalities. METHODS: All BRCA1/2 patients at a single academic center were retrospectively reviewed (2005-2015). Pancreatic abnormalities were defined on cross-sectional imaging as pancreatic neoplasm (cystic/solid) or main-duct dilation. RESULTS: Two hundred and four patients were identified with BRCA mutations. Forty-seven (40%) had abdominal imaging (20 computerized tomography and 27 magnetic resonance imaging). Twenty-one percent had pancreatic abnormalities (PDAC [n = 2] and intraductal papillary mucinous neoplasm [IPMN; n = 8]). The prevalence of pancreatic abnormalities and IPMN was higher in BRCA2 patients than in the general population (21% vs 8% and 17% vs 1%; P = 0.0007 and P < 0.0001, respectively), with no influence of family history. Similarly, BRCA1 patients had an increased prevalence of IPMN (8.3% vs 1%; P < 0.0001). CONCLUSIONS: In this series, 4% and 17% of BRCA2 patients developed PDAC and IPMN, respectively. Eight percent of BRCA1 patients developed IPMN. Under current recommended screening, 60% of BRCA1/2 patients had incompletely pancreatic assessment. With no influence of family history, this study suggests all BRCA1/2 patients should undergo a high-risk screening protocol that will identify a higher rate of precancerous pancreatic neoplasms amenable to curative resection.

Patient Selection for Total Pancreatectomy with Islet Autotransplantation in the Surgical Management of Chronic Pancreatitis.

BACKGROUND: Best practice to select patients with chronic pancreatitis for surgical management with total pancreatectomy with islet autotransplantation (TPIAT) is in evolution as new discoveries are made in the pathogenesis of chronic pancreatitis. STUDY DESIGN: A prospectively maintained database of patients undergoing TPIAT was reviewed. Islet function was inferred from daily insulin requirement. Pain relief was evaluated by healthcare use and narcotic use. Quality of life (QOL) was measured with the RAND 12-Item Short Form Survey. RESULTS: One hundred and ninety-five patients (141 women, aged 40.3 years, BMI 26.5 kg/m2) underwent TPIAT. Mean duration of disease before operation was 8.1 years. Fifty-six (29%) patients had pancreatic operations before TPIAT, 37 (19%) patients were diabetic preoperatively, and 52 (27%) patients were smokers. A mean of 3,253 islet equivalents transplanted/kg were harvested. Insulin independence was achieved in 29%, 28%, and 23% of patients at 1, 2, and 5 years postoperative. Nonsmokers with a shorter duration of chronic pancreatitis and no earlier pancreas operation were more likely to be insulin free. Median number of preoperative emergency department visits and hospitalizations were 6.6 and 4.3 annually, respectively, compared with 0 at 1, 2, and 5 years postoperative. Median oral morphine equivalents were 214 mg/kg preoperation and 60, 64, 69, at 1, 2, 5 years postoperative. Preoperative, 1, 2, 5 years postoperative QOL scores were 29, 36, 34, and 33 (physical; p < 0.01) and 39, 44, 42, and 42 (mental health; p < 0.02). Genetic pancreatitis patients were more often narcotic free and had better QOL than patients with pancreatitis of other causes. At 5 years, overall survival was 92.3%. CONCLUSIONS: Total pancreatectomy with islet autotransplantation is a durable operation, with islet function, pain relief, and QOL improvements persisting to 5 years postoperative. Patients with genetic pancreatitis, short duration of disease, and nonsmokers have superior outcomes.

Indication for en bloc pancreatectomy with colectomy: when is it safe?

INTRODUCTION: Aggressive en bloc resection of adjacent organs is often necessary to resect pancreatic or colonic lesions. However, it is debated whether simultaneous pancreatectomy with colectomy (P+C) is warranted as it potentially increases morbidity and mortality (MM). We hypothesized that MM would be increased in P+C, especially in cases of pancreatitis. METHODS: All patients who underwent pancreatectomy (P) and simultaneous pancreatectomy with colectomy (P+C) at a high-volume center from November 2006 to 2015 were prospectively collected using ACS-NSQIP at our institution. Patients with additional multivisceral or enucleation procedures were excluded. Data were augmented to 90-day outcomes using our institutional database. RESULTS: Forty-three patients with a mean age of 62 years (27:16 male: female) underwent P+C, accounting for 2.39% (43/1797) of pancreatectomies performed. Pancreatoduodenectomy (PD) was performed in 61% (n = 26), distal pancreatectomy (DP) in 37% (n = 16), and total pancreatectomy (TP) in 2% (n = 1) of patients. The 30- and 90-day MM were higher in P+C than P (30-day: 54 vs. 37%, p = 0.037 and 9 vs. 2%, p = 0.022; 90-day: 61 vs. 42%, p = 0.019 and 14 vs. 3%, p = 0.002). Logistical regression modeling revealed an association between 90-day mortality and colectomy (p = 0.013, OR = 3.556). When P+C MM were analyzed according to intraoperative factors, there was no significant difference according to type of pancreatectomy (PD vs. DP vs. TP), origin of primary lesion (pancreas vs. colon), surgical indication (malignant vs. non-malignant), or case status (planned colectomy vs. intraoperative decision). CONCLUSIONS: Addition of colectomy to pancreatectomy substantially increased MM. Subanalysis revealed that type of resection performed, etiology, and planning status did not account for increased risk when performing P+C. However, colectomy was found to be an independent risk factor for mortality. Therefore, patients should be informed of the risk of increased postoperative complications until a further study can identify potential patients or perioperative factors that can be used for risk stratification.

Enhanced Recovery After Surgery Protocols Are Valuable in Pancreas Surgery Patients.

BACKGROUND: There is increasing interest in implementing comprehensive perioperative protocols, including preoperative optimization and education, perioperative goal-directed fluid management, and postoperative fast tracking, to enhance recovery after surgery. Data on the outcomes of these protocols in pancreatic surgery, however, are limited. STUDY DESIGN: A retrospective review of a prospectively maintained pancreas surgery database at a single institution from August 2012 to April 2015 was undertaken. An enhanced recovery protocol was initiated in October 2014, and patients were divided into groups according to preprotocol or postprotocol implementation. Preoperative, intraoperative, and postoperative data were tabulated. Statistical analysis was performed with Student's t-test and Fisher's exact tests, as well as equality of variances where appropriate, using SAS System software (SAS Institute). RESULTS: Three hundred and seventy-eight patients (181 men, mean age 54 years, BMI 28 kg/m(2)) underwent elective pancreatic surgery during the study period, 297 patients preprotocol and 81 postprotocol. There were no significant differences in preoperative or intraoperative characteristics. Mean postoperative length of stay was significantly lower in the Enhanced Recovery After Surgery group (7.4 vs 9.2 days; p < 0.0001). Hospital costs were similarly lower in the Enhanced Recovery After Surgery group ($23,307.90 vs $27,387.80; p < 0.0001). Readmission (29% vs 32%) and pancreatic fistula (26% vs 28%) rates were similar between groups. Delayed gastric emptying was lower in the Enhanced Recovery After Surgery group (26% vs 13%; p = 0.03). CONCLUSIONS: Implementation of an enhanced recovery after pancreatic surgery protocol significantly decreased length of stay and hospital cost without increasing readmission or morbidity. Despite patient complexity and the potential need for individualization of care, enhanced recovery protocols can be valuable and effective in high-risk patient populations, including pancreatic surgery patients.

Use of recombinant bone morphogenetic protein 2 in free flap reconstruction for osteonecrosis of the mandible.

PURPOSE: Osteoradionecrosis of the mandible is a debilitating consequence of radiation therapy for head-and-neck malignancy. It can result in pain, bone exposure, fistula formation, and pathologic fracture. Recombinant human bone morphogenetic protein 2 (rhBMP-2) has shown promise in reconstruction of bone defects. The purpose of this study is to determine whether the addition of rhBMP-2 at the union of vascularized bone and native bone improves surgical outcomes in patients with osteonecrosis of the mandible. MATERIALS AND METHODS: This study was a retrospective analysis of patients who were treated between 2006 and 2010 for osteonecrosis of the mandible. Patients requiring definitive reconstruction after failure of a course of conservative management were included. Patients were divided into 2 cohorts depending on whether rhBMP-2 was used during the reconstruction. The primary outcome measure was defined as stable mandibular union. RESULTS: Seventeen patients were included. The development of malunion was similar in both groups (13% for rhBMP-2 group vs 11% for non-rhBMP-2 group). Infectious complications were similar between the groups (25% in rhBMP-2 group vs 56% in non-rhBMP-2 group, P = .33). The rates of hardware removal were similar for the 2 groups (33% in non-rhBMP-2 group vs 25% in rhBMP-2 group, P = .10). No cancer recurrences were observed in patients receiving rhBMP-2. CONCLUSIONS: The use of rhBMP-2 is safe in free flap reconstruction of the mandible, but its ability to significantly improve patient outcomes, as measured by rates of malunion, reoperation, or infection, is still unknown.

Outcomes of salvage surgery with free flap reconstruction for recurrent oral and oropharyngeal cancer.

OBJECTIVES/HYPOTHESIS: To evaluate outcomes of salvage surgery with free flap reconstruction for recurrent squamous cell carcinoma of the oropharynx and oral cavity with increased use of chemoradiotherapy. STUDY DESIGN: Retrospective patient review. METHODS: All patients undergoing salvage surgery with free flap reconstruction for oropharynx (n = 36) and oral cavity (n = 36) squamous cell carcinomas between January 2001 and January 2008 were obtained. Mean follow-up was 14 months. Previous chemoradiotherapy was used in 40% and radiotherapy alone in 60%. RESULTS: Complications were more frequent in oropharynx than oral cavity tumors (36% and 14%, respectively; P = .05) requiring more secondary procedures (15 for oropharynx vs. six for oral cavity). Few patients returned to a normal diet (8%), and a majority retained an enterogastric feeding tube (56%). Median survival overall following salvage surgery was 44.8 months for oral cavity and 53.8 months for oropharynx head and neck squamous cell carcinoma. Overall estimated 1-, 2-, and 5-year observed survivals were 98%, 77.2%, and 43.7%, respectively. Twelve patients had a disease-free interval of <6 months, 92% of whom died of disease. Of 17 patients with disease at the primary site and involved regional lymph nodes, 94% died of disease. CONCLUSIONS: Salvage surgery with free flap reconstruction for recurrent oral and oropharyngeal tumors after chemoradiotherapy has acceptable morbidity and similar cure rates as salvage following radiotherapy without chemotherapy. Concurrent nodal recurrence and short disease-free interval are associated with reduced cure rates. A significant proportion will require enterogastric feeding and few will tolerate a normal diet.