Warren H. Tseng

Surgical Resident Involvement Is Safe for Common Elective General Surgery Procedures

BACKGROUND Outcomes of surgical resident training are under scrutiny with the changing milieu of surgical education. Few have investigated the effect of surgical resident involvement (SRI) on operative parameters. Examining 7 common general surgery procedures, we evaluated the effect of SRI on perioperative morbidity and mortality and operative time (OpT). STUDY DESIGN The American College of Surgeons National Surgical Quality Improvement Program database (2005 to 2007) was used to identify 7 cases of nonemergent operations. Cases with simultaneous procedures were excluded. Logistic regression was performed across all procedures and within each procedure incorporating SRI, OpT, and risk-stratifying American College of Surgery National Surgical Quality Improvement Program morbidity and mortality probability scores, which incorporate multiple prognostic individual patient factors. Procedure-specific, SRI-stratified OpTs were compared using Wilcoxon rank-sum tests. RESULTS A total of 71.3% of the 37,907 cases had SRI. Absolute 30-day morbidity for all cases with SRI and without SRI were 3.0% and 1.0%, respectively (p < 0.001); absolute 30-day mortality for all cases with SRI and without SRI were 0.1% and 0.08%, respectively (p < 0.001). After multivariate analysis by specific procedure, SRI was not associated with increased morbidity but was associated with decreased mortality during open right colectomy (odds ratio 0.32; p = 0.01). Across all procedures, SRI was associated with increased morbidity (odds ratio 1.14; p = 0.048) but decreased mortality (odds ratio 0.42; p < 0.001). Mean OpT for all procedures was consistently lower for cases without SRI. CONCLUSIONS SRI has a measurable impact on both 30-day morbidity and mortality and OpT. These data have implications to the impact associated with surgical graduate medical education. Further studies to identify causes of patient morbidity and prevention strategies in surgical teaching environments are warranted.

Tumor Location Predicts Survival in Cutaneous Head and Neck Melanoma

BACKGROUND Prior studies documented poorer outcomes in patients with cutaneous head and neck melanoma (CHNM) relative to those with melanoma at other sites. We evaluated survival differences attributable to tumor location in patients with CHNM. METHODS We queried the Surveillance, Epidemiology, and End Results (SEER) database for patients undergoing surgery for CHNM from 1988 to 2006, excluding patients without biopsy-proven diagnoses, those diagnosed at autopsy, and patients with distant metastases. Using the Kaplan-Meier method, we assessed patient, tumor, and treatment-specific factors on overall survival (OS) and melanoma specific survival (MSS). Cox proportional hazards models assessed the role of tumor location (ear, eyelid, face, lip, scalp/neck) on OS and MSS, while controlling for patient age, gender, race, tumor thickness, tumor ulceration, lymph node status, histologic subtype, type of surgery, and use of radiation. Risks of overall and melanoma-specific mortality were reported as hazard ratios (HR) with 95% confidence intervals (CI). RESULTS Among 27,097 patients, 10-y rates of OS and MSS were 56.1% and 84.7%, respectively. On multivariate analysis, scalp/neck primary site was associated with an increased risk of overall (HR 1.20, CI 1.14-1.26; P < 0.001) and melanoma-specific mortality (HR 1.64, CI 1.49-1.80, P < 0.001) relative to melanomas of the face. Tumors of the lip had poorer MSS (HR 1.55; CI 1.05-2.28, P = 0.03) but not OS (HR 1.03, CI 0.80-1.34; P = 0.80). CONCLUSIONS Patients with melanomas of the scalp/neck have poorer OS and MSS and those with lip melanomas have poorer MSS. These anatomic areas should not be overlooked when performing skin examinations.

Predicting Survival for Well-Differentiated Liposarcoma: The Importance of Tumor Location

BACKGROUND Although well-differentiated liposarcoma (WD Lipo) is a low grade neoplasm with a negligible risk of metastatic disease, it can be locally aggressive. We hypothesized that survival for WD Lipo varies significantly based on tumor location. METHODS We identified 1266 patients with WD Lipo in the Surveillance, Epidemiology, and End Results database from 1988-2004. After excluding patients diagnosed by autopsy only, those lacking histologic confirmation, those lacking data on tumor location, and those with metastatic disease or unknown staging information, we arrived at a final study cohort of 1130 patients. Clinical, pathologic, and treatment variables were analyzed for their association with overall survival (OS) and disease-specific survival (DSS) using Kaplan-Meier analysis and Cox proportional hazards multivariate models. RESULTS Mean age was 61 y (± 14.6), 72.2% were white, and 60.4% were male. Eighty-one percent of patients were treated with surgical therapy alone, 4.6% were treated with radiotherapy (RT) alone, and 12.9% were treated with both surgery and RT. Extremity location was most common (41.6%), followed by trunk (29%), retroperitoneal/intra-abdominal (RIA, 21.6%), thorax (4.2%), and head/neck (3.6%). With a median follow-up of 45 mo, median OS was 115 mo (95% confidence interval [CI] 92-138 mo) for RIA tumors compared to not reached for other tumor locations (P = 0.002). On multivariate analysis, increasing age and RIA location both predicted worse OS and DSS while tumor size, race, sex, receipt of RT, and Surveillance, Epidemiology, and End Results (SEER) stage did not. Tumor size became a significant predictor of worse DSS, but not OS, only when site, SEER stage, and extent of resection were removed from the multivariate model. Non-RIA locations, including extremity, experienced statistically similar OS, but 5-y DSS for trunk location was intermediate [92.3%, (95% CI 88.5%-96.1%) compared with 98.0% (95% CI, 96.2%-99.8%) for extremity and 86.6 (95% CI 81.1%-92.1%) for RIA, P < 0.001]. CONCLUSIONS Among patients with WD Lipo, RIA location is associated with significantly worse outcomes independent of tumor size. Future studies should focus on the anatomic and biologic reasons for these differences.

Lack of survival benefit following adjuvant radiation in patients with retroperitoneal sarcoma: A SEER analysis

BACKGROUND The benefit of radiation therapy (RT) among patients with retroperitoneal sarcoma (RPS) is controversial. We performed a retrospective analysis of the effect of RT on survival among RPS patients using a nationwide cancer registry. METHODS Utilizing data from the Surveillance, Epidemiology, and End Results (SEER) database, we identified 2308 cases of RPS from 1988 to 2004. We excluded 773 cases for age < 18, identification by autopsy only, absence of histologic confirmation, presence of metastatic disease, or lack of surgical intervention. Overall survival (OS) and disease-specific survival (DSS) were estimated using the Kaplan-Meier method. Multivariate analysis was performed using a Cox proportional hazards model, adjusting for significant covariables. RESULTS Among 1535 patients who met entry criteria, RT was administered to 373 patients (24.3%). The majority of RT (n = 300, 80.4%) was administered postoperatively. Median OS was 60 and 60 mo, respectively, for patients receiving and not receiving RT (P = 0.59). Median DSS was 86 and 117 mo, respectively, for patients receiving and not receiving RT (P = 0.84). On multivariate analysis, younger age, female gender, low and intermediate histologic grade, liposarcoma histology, tumor size 5-10 cm, and completeness of resection all independently predicted better OS and DSS, while RT did not (HR for OS with RT 0.92, 95% CI 0.78-1.09 and HR for DSS with RT 0.96, 95% CI 0.78-1.17). On subgroup analysis by histology, patients with malignant fibrous histiocytoma (MFH) receiving RT demonstrated statistically improved OS (P = 0.002) and DSS (P = 0.01), respectively. CONCLUSIONS With the possible exception of MFH, postoperative RT offers no survival benefit in RPS. Further studies are necessary to determine if the selective application of RT is indicated.

Treatment Options for Metaplastic Breast Cancer

Metaplastic breast cancer (MBC) is a malignancy characterized by the histologic presence of two or more cellular types, commonly a mixture of epithelial and mesenchymal components. MBC is rare relative to invasive ductal carcinoma (IDC), representing less than 1% of all breast cancers. Other than a lower rate of lymph node metastases, MBC tumors display poorer prognostic features relative to IDC. Due to its low incidence and pathological variability, the ideal treatment paradigm for MBC is unknown. Because of its rarity, MBC has been treated as a variant of IDC. Despite similar treatment regimens, however, patients with MBC have worse outcomes. Recent research is focused on biological differences between MBC and IDC and potential novel targets for chemotherapeutic agents. This paper serves as a summation of current literature on approaches to the multidisciplinary treatment of patients with MBC.

Contiguous organ resection is safe in patients with retroperitoneal sarcoma: An ACS-NSQIP analysis.

The practice of aggressive contiguous organ resection (COR) of retroperitoneal sarcoma (RPS) is controversial. We examined rates of 30‐day morbidity and mortality following resection of RPS utilizing data from the American College of Surgeons National Surgical Quality Improvement Program (ACS‐NSQIP) database.

Nomogram to predict risk of 30-day morbidity and mortality for patients with disseminated malignancy undergoing surgical intervention.

Objective: To estimate individual risk of 30-day surgical morbidity and mortality after surgical intervention for patients with disseminated malignancy (DMa). Background: Patients with DMa frequently require surgical consultation for palliative operations. Although these patients are at high risk for surgical morbidity and mortality, limited data exist allowing individual risk stratification. Methods: Using the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) from 2005 to 2007, we identified 7447 patients with DMa. Each of the 53 preoperative ACS NSQIP variables was analyzed to assess risk of morbidity and mortality. Logistic regression models were developed using stepwise model selection and generalized additive models. Covariates were evaluated for nonlinearity and interactions among variables. We constructed nomograms utilizing clinically and statistically significant covariates to predict 30-day risk of morbidity and mortality. Results: Overall 30-day unadjusted morbidity and mortality rates were 28.3% and 8.9%, respectively. Mortality rates reached 18.4% for vascular procedures and 27.9% for emergent operations. Increasing age, impaired functional status, Do-Not-Resuscitate status, impaired respiratory function, ascites, hypoalbuminema, elevated creatinine, and abnormal WBC were all significant predictors (P < 0.0001) of increased morbidity and mortality on multivariate analysis. Nomograms to predict individual 30-day risk of complications and death based on preoperative factors were developed and validated by bootstrapping. Concordance indices were 0.704 and 0.861 for morbidity and mortality, respectively. Conclusions: Surgical intervention among patients with DMa is associated with substantial morbidity and mortality. We have constructed nomograms to predict individual risk of 30-day morbidity and mortality. These have significant implications for surgical decision-making in this group of patients.

Sacramento area breast cancer epidemiology study: use of postmastectomy breast reconstruction along the rural-to-urban continuum.

Background: Health care disparities have been documented in rural populations. The authors hypothesized that breast cancer patients in urban counties would have higher rates of postmastectomy breast reconstruction relative to patients in surrounding near-metro and rural counties. Methods: The authors used the Surveillance, Epidemiology, and End Results database to identify patients diagnosed with breast cancer and treated with mastectomy in the greater Sacramento area between 2000 and 2006. Counties were categorized as urban, near-metro, or rural. Univariate models evaluated the relationship of rural, near-metro, or urban location with use of breast reconstruction by means of the chi-square test. Multivariate logistic regression models controlling for patient, tumor, and treatment-related factors predicted use of breast reconstruction. The likelihood of undergoing breast reconstruction was reported as odds ratios with 95 percent confidence intervals; significance was set at p ≤ 0.05. Results: Complete information was available for 3552 breast cancer patients treated with mastectomy. Of these, 718 (20.2 percent) underwent breast reconstruction. On univariate analysis, differences in the rates of breast reconstruction were noted among urban, near-metro, and rural areas (p < 0.001). On multivariate analysis, patients from rural (odds ratio, 0.51; 95 percent confidence interval, 0.28 to 0.93; p < 0.03) and near-metro (odds ratio, 0.73; 95 percent confidence interval, 0.59 to 0.89; p = 0.002) areas had a decreased likelihood of undergoing breast reconstruction relative to patients from urban areas. Conclusions: Patients from near-metro and rural areas are less likely to undergo breast reconstruction following mastectomy for breast cancer than their urban counterparts. Differences in use of breast reconstruction detected at a population level should guide future interventions to increase rates of breast reconstruction at the local level.

Do radiation use disparities influence survival in patients with advanced breast cancer

The authors previously identified racial/ethnic disparities in the use of radiation therapy (RT) in patients with advanced breast cancer (BC). They hypothesized that disparities in the use of RT were associated with survival differences favoring white patients.