Using risk-adjusted cumulative sum to evaluate surgeon, divisional, and institutional outcomes-a feasibility study.

BACKGROUND: Few objective, real-time measurements of surgeon performance exist. The risk-adjusted cumulative sum is a novel method that can track surgeon-level outcomes on a continuous basis. The objective of this study was to demonstrate the feasibility of using risk-adjusted cumulative sum to monitor outcomes after colorectal operations and identify clinically relevant performance variations. METHODS: The National Surgical Quality Improvement Program was queried to obtain patient-level data for 1,603 colorectal operations at a high-volume center from 2011 to 2020. For each case, expected risks of morbidity, mortality, reoperation, readmission, and prolonged length of stay were estimated using the National Surgical Quality Improvement Program risk calculator. Risk-adjusted cumulative sum curves were generated to signal observed-to-expected odds ratios of 1.5 (poor performance) and 0.5 (exceptional performance). Control limits were set based on a false positive rate of 5% (α = 0.05). RESULTS: The cohort included data on 7 surgeons (those with more than 20 cases in the study period). Institutional observed versus expected outcomes were the following: morbidity 12.5% (vs 15.0%), mortality 2.5% (vs 2.0%), prolonged length of stay 19.7% (vs 19.1%), reoperation 11.1% (vs 11.3%), and 30-day readmission 6.1% (vs 4.8%). Risk-adjusted cumulative sum accurately demonstrated within- and between-surgeon performance variations across these metrics and proved effective when considering division-level data. CONCLUSION: Risk-adjusted cumulative sum adjusts for patient-level risk factors to provide real-time data on surgeon-specific outcomes. This approach enables prompt identification of performance outliers and can contribute to quality assurance, root-cause analysis, and incentivization not only at the surgeon level but at divisional and institutional levels as well.

Laparoscopic surgery for adrenocortical carcinoma: Estimating the risk of margin-positive resection.

BACKGROUND: Over recent years, there has been increasing adoption of minimally invasive surgery (MIS) in the treatment of adrenocortical carcinoma (ACC). However, MIS has been associated with noncurative resection and locoregional recurrence. We aimed to identify risk factors for margin-positivity among patients who undergo MIS resection for ACC. We hypothesized that a simple nomogram can accurately identify patients most suitable for curative MIS resection. METHODS: Curative-intent resections for ACC were identified through the National Cancer Database spanning 2010-2018. Trends in MIS utilization were reported using Pearson correlation coefficients. Factors associated with margin-positive resection were identified among preoperatively available variables using multivariable logistic regression, then incorporated into a predictive model. Model quality was cross validated using an 80% training data set and 20% test data set. RESULTS: Among 1260 ACC cases, 38.6% (486) underwent MIS resection. MIS utilization increased over time at nonacademic centers (R = 0.818, p = 0.007), but not at academic centers (R = 0.009, p = 0.982). Factors associated with margin-positive MIS resection were increasing age, nonacademic center (odds ratio [OR]: 1.8, p = 0.006), cT3 (OR: 4.7, p < 0.001) or cT4 tumors (OR: 14.6, p < 0.001), and right-sided tumors (OR: 2.0, p = 0.006). A predictive model incorporating these four factors produced favorable c-statistics of 0.75 in the training data set and 0.72 in the test data set. A pragmatic nomogram was created to enable bedside risk stratification. CONCLUSIONS: An increasing proportion of ACC are resected via minimally invasive operations, particularly at nonacademic centers. Patient selection based on a few key factors can minimize the risk of noncurative surgery.

Age-stratified comparison of active surveillance versus radiofrequency ablation for papillary thyroid microcarcinoma using decision analysis.

BACKGROUND: Papillary thyroid microcarcinomas may be treated with radiofrequency ablation, active surveillance, or surgery. The objective of this study was to use mathematical modeling to compare treatment alternatives for papillary thyroid microcarcinomas among those who decline surgery. We hypothesized that radiofrequency ablation would outperform active surveillance in avoiding progression and surgery but that the effect size would be small for older patients. METHODS: We engaged stakeholders to identify meaningful long-term endpoints for papillary thyroid microcarcinoma treatment-(1) cancer progression/surgery, (2) need for thyroid replacement therapy, and (3) permanent treatment complication. A Markov decision analysis model was created to compare the probability of these endpoints after radiofrequency ablation or active surveillance for papillary thyroid microcarcinomas and overall cost. Transition probabilities were extracted from published literature. Model outcomes were estimated to have a 10-year time horizon. RESULTS: The primary outcome yielded a number needed to treat of 18.1 for the avoidance of progression and 27.4 for the avoidance of lifelong thyroid replacement therapy for radiofrequency ablation compared to active surveillance. However, as patient age increased, the number needed to treat to avoid progression increased from 5.2 (age 20-29) to 39.1 (age 60+). The number needed to treat to avoid lifelong thyroid replacement therapy increased with age from 7.8 (age 20-29) to 59.3 (age 60+). The average 10-year cost/treatment for active surveillance and radiofrequency ablation were $6,400 and $11,700, respectively, translating to a cost per progression-avoided of $106,500. CONCLUSION: As an alternative to active surveillance, radiofrequency ablation may have a greater therapeutic impact in younger patients. However, routine implementation may be cost-prohibitive for most patients with papillary thyroid microcarcinomas.

Outcomes after endoscopic local excision of early-stage gastric adenocarcinoma in the United States.

BACKGROUND AND OBJECTIVES: Local excision (LE) for early-stage gastric cancer has expanded in the United States over recent years, however, national outcomes are unknown. The objective of the study was to evaluate national survival outcomes following LE for early-stage gastric cancer. METHODS: Patients with resectable gastric adenocarcinoma between 2010 and 2016 were identified from the National Cancer Database then classified by LE curability into eCuraA (high) and eCuraC (low) according to Japanese Gastric Cancer Association guidelines. Demographics, clinical/provider descriptors, and perioperative/survival outcomes were extracted. Propensity-weighted cox proportional hazards regression assessed factors associated with overall survival. RESULTS: Patients were stratified into eCuraA (N = 1167) and eCuraC (N = 13,905) subgroups. Postoperative 30-day mortality (0% vs 2.8%, p < 0.001) and readmission (2.3% vs 7.8%, p = 0.005) favored LE. Local excision was not associated with survival on propensity-weighted analyses. However, among eCuraC patients, LE was associated with higher likelihood of positive margins (27.1% vs 7.0%, p < 0.001), which was the strongest predictor of poor survival (HR 2.0, p < 0.001). CONCLUSIONS: Although early morbidity is low, oncologic outcomes following LE are compromised for eCuraC patients. These findings support careful patient selection and treatment centralization in the early adoption phase of LE for gastric cancer.

Eliciting Low-Risk Thyroid Cancer Treatment Preferences Using Clinical Vignettes: A Pilot Study.

OBJECTIVE: While surgical resection has been the traditional standard treatment for small (≤1 cm), differentiated thyroid cancers, active surveillance (AS) and radiofrequency ablation (RFA) are increasingly considered. The aim of this study was to explore patient preferences in thyroid cancer treatment using a series of clinical vignettes. METHODS: Thyroid cancer survivors and general population volunteers were recruited to rank experience-driven clinical vignettes in order of preference. Rankings were compared using Wilcoxon signed rank. Formative qualitative methods were used to develop and refine clinical vignettes that captured 4 treatments-thyroid lobectomy (TL), total thyroidectomy (TT), AS, and RFA-along with 6 treatment complications. Content was validated via interviews with 5 academic subspecialists. RESULTS: Nineteen volunteers participated (10 survivors, 9 general population). Treatment complications were ranked lower than uncomplicated counterparts in 99.0% of cases, indicating excellent comprehension. Counter to our hypothesis, among uncomplicated vignettes, median rankings were 1 for AS, 2 for RFA, 3.5 for TL, and 5 for TT. Trends were consistent between thyroid cancer survivors and the general population. AS was significantly preferred over RFA (P = .02) and TT (P < .01). Among surgical options, TL was significantly preferred over TT (P < .01). CONCLUSION: When treatments for low-risk thyroid cancer are described clearly and accurately through clinical vignettes, patients may be more likely to choose less invasive treatment options over traditional surgical resection.

Monitoring performance in laparoscopic gastric bypass surgery using risk-adjusted cumulative sum at 2 high-volume centers.

BACKGROUND: Traditional surgical outcomes are measured retrospectively and intermittently, limiting opportunities for early intervention. OBJECTIVES: The objective of this study was to use risk-adjusted cumulative sum (RA-CUSUM) to track perioperative surgical outcomes for laparoscopic gastric bypass. We hypothesized that RA-CUSUM could identify performance variations between surgeons. SETTING: Two mid-Atlantic quaternary care academic centers. METHODS: Patient-level data from the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program (MBSAQIP) were abstracted for laparoscopic gastric bypasses performed by 3 surgeons at 2 high-volume centers from 2014 to 2021. Estimated probabilities of serious complications, reoperation, and readmission were derived from the MBSAQIP risk calculator. RA-CUSUM curves were generated to signal observed-to-expected odds ratios (ORs) of 1.5 (poor performance) and .5 (superior performance). Control limits were set based on a false positive rate of 5% (α = .05). RESULTS: We included 1192 patients: Surgeon A = 767, Surgeon B = 188, and Surgeon C = 237. Overall rates of serious complications, 30-day reoperations, and 30-day readmissions were 3.9%, 2.5%, and 5.2% respectively, with expected rates of 4.7%, 2.2%, and 5.8%. RA-CUSUM signaled lower-than-expected (OR < .5) rates of readmission and serious complication in Surgeon A, and higher-than-expected (OR > 1.5) readmission rate in Surgeon C. Surgeon A further demonstrated an early period of higher-than-expected (OR > 1.5) reoperation rate before April 2015, followed by superior performance thereafter (OR < .5). Surgeon B's performance generally reflected expected standards throughout the study period. CONCLUSIONS: RA-CUSUM adjusts for clinical risk factors and identifies performance outliers in real-time. This approach to analyzing surgical outcomes is applicable to quality improvement, root-cause analysis, and surgeon incentivization.

If the price is right: Cost-effectiveness of radiofrequency ablation versus thyroidectomy in the treatment of benign thyroid nodules.

BACKGROUND: Radiofrequency ablation is an emerging technology in the United States to treat benign thyroid nodules. The cost-effectiveness of radiofrequency ablation in comparison with traditional thyroidectomy is unknown. METHODS: A patient-level state transition microsimulation decision model was constructed comparing radiofrequency ablation with lobectomy in the management of benign thyroid nodules. Our base case was a 45-year-old woman with a solitary 30-cm3 nodule. Estimates of health utilities, complications, and mortality were obtained from the literature, and costs were estimated using Medicare reimbursement data. The primary outcomes of interest included total cost, quality-adjusted life years, and incremental cost-effectiveness ratios. All model estimates were subjected to 1-way sensitivity analyses to identify factors that strongly influence cost-effectiveness. A probabilistic sensitivity analysis was run across 1 million simulations to gauge outcome confidence with a willingness-to-pay threshold set at $100,000/quality-adjusted life year. RESULTS: Radiofrequency ablation was assumed to cost $5,000, with an initial success rate of 78%. Patients with volume reduction ratio <50% underwent a second treatment of radiofrequency ablation. Radiofrequency ablation represented the dominant strategy, yielding 21.31 quality-adjusted life years for a total cost of $16,563 in comparison to lobectomy, which yielded 21.13 quality-adjusted life years for a total cost of $19,262. In a 1-way sensitivity analysis varying the cost of radiofrequency ablation across of range of values, the radiofrequency ablation strategy remained cost-effective until the cost of radiofrequency ablation exceeded $12,330 at willingness-to-pay $50,000 or $17,950 at willingness-to-pay $100,000. CONCLUSION: Radiofrequency ablation is a cost-effective strategy in the treatment of benign thyroid nodules but is most sensitive to the cost of radiofrequency ablation.

A Novel Thyroid Ultrasound Proficiency Metric Designed Through a Multidisciplinary Delphi Approach.

BACKGROUND: Thyroid ultrasounds extend surgeons' outpatient capabilities and are essential for operative planning. However, most residents are not formally trained in thyroid ultrasound. The purpose of this study was to create a novel thyroid ultrasound proficiency metric through a collaborative Delphi approach. METHODS: Clinical faculty experienced in thyroid ultrasound participated on a Delphi panel to design the thyroid Ultrasound Proficiency Scale (UPS-Thyroid). Participants proposed items under the categories of Positioning, Technique, Image Capture, Measurement, and Interpretation. In subsequent rounds, participants voted to retain, revise, or exclude each item. The process continued until all items had greater than 70% consensus for retention. The UPS-Thyroid was pilot tested across 5 surgery residents with moderate ultrasound experience. Learning curves were assessed with cumulative sum. RESULTS: Three surgeons and 4 radiologists participated on the Delphi panel. Following 3 iterative Delphi rounds, the panel arrived at >70% consensus to retain 14 items without further revisions or additions. The metric included the following items on a 3-point scale for a maximum of 42 points: Positioning (1 item), Technique (4 items), Image Capture (2 items), Measurement (2 items), and Interpretation (5 items). A pilot group of 5 residents was scored against a proficiency threshold of 36 points. Learning curve inflection points were noted at between 4 to 7 repetitions. CONCLUSIONS: A multidisciplinary Delphi approach generated consensus for a thyroid ultrasound proficiency metric (UPS-Thyroid). Among surgery residents with moderate ultrasound experience, basic proficiency at thyroid ultrasound is feasible within 10 repetitions.

Regional control after precision lymph node dissection for clinically evident melanoma metastasis.

INTRODUCTION: Completion lymph node dissection (CLND) for microscopic lymph node metastases has been replaced by observation; however, CLND is standard for clinically detectable nodal metastases (cLN). CLND has high morbidity, which may be reduced by excision of only the cLN (precision lymph node dissection [PLND]). We hypothesized that same-basin recurrence risk would be low after PLND. METHODS: Retrospective review at four tertiary care hospitals identified patients who underwent PLND. The primary outcome was 3-year cumulative incidence of isolated same-basin recurrence. RESULTS: Twenty-one patients underwent PLND for cLN without synchronous distant metastases. Reasons for forgoing CLND included patient preference (n = 11), comorbidities (n = 5), imaging indeterminate for distant metastases (n = 2), partial response to checkpoint blockade (n = 1), or not reported (n = 2). A median of 2 nodes (range: 1-6) were resected at PLND, and 68% contained melanoma. Recurrence was observed in 33% overall. Only 1 patient (5%) developed an isolated same-basin recurrence. Cumulative incidences at 3 years were 5.0%, 17.3%, and 49.7% for isolated same-basin recurrence, any same-basin recurrence, and any recurrence, respectively. Complications from PLND were reported in 1 patient (5%). CONCLUSIONS: These pilot data suggest that PLND may provide adequate regional disease control with less morbidity than CLND. These data justify prospective evaluation of PLND in select patients.

Total thyroidectomy is more cost-effective than radioactive iodine as an alternative to antithyroid medication for Graves' disease.

BACKGROUND: Patients with Graves' disease treated with radioactive iodine report worse quality of life than those treated by thyroidectomy. However, radioactive iodine is often selected due to lower risk of complications and lower cost. The objective of this study was to estimate the cost-effectiveness of radioactive iodine versus total thyroidectomy for treatment of Graves' disease. METHODS: A Markov decision-analytic model was created to simulate clinical outcomes and costs of medication-refractory Graves' disease treated with radioactive iodine or total thyroidectomy. Complication rates and utilities were derived from published data. Costs were extracted from national Medicare reimbursement rates. We conducted 1-way, 2-way, and probabilistic sensitivity analyses to identify factors that influence cost-effectiveness and reflect uncertainty in model parameters. The willingness-to-pay threshold was set at $100,000/quality-adjusted life-years. RESULTS: Total thyroidectomy yielded 23.6 quality-adjusted life-years versus 20.9 quality-adjusted life-years for radioactive iodine. The incremental cost-effectiveness ratio was $2,982 per quality-adjusted life-years, indicating that surgery is highly cost-effective relative to radioactive iodine. Surgery was more cost effective than radioactive iodine in 88.2% of model simulations. Sensitivity analyses indicate that the model outcomes are driven predominantly by posttreatment quality of life, with contributing effects from rates of treatment complications and the impact of these complications on quality of life. CONCLUSION: For patients with Graves' disease who either cannot tolerate or are refractory to antithyroid drugs, thyroidectomy is more cost-effective than radioactive iodine. Future research should validate reported differences in quality of life between these 2 treatment modalities.

Constructing Health State Descriptions for Low-Risk Thyroid Cancer: Stakeholder Engagement and Formative Qualitative Research.

OVERVIEW: This paper describes stakeholder involvement and formative qualitative research in the creation of health state descriptions (HSDs) or vignettes for low-risk thyroid cancer. The aim of this project was to engage stakeholders in the contribution of a novel set of HSDs, an important first step in the process of assessing value in thyroid cancer health states. METHODS: We draw upon formative, descriptive qualitative methods, following a multi-stage framework of data collection. We conducted individual semi-structured interviews, cognitive interviews, and focus groups with thyroid cancer patients, community providers, academic subspecialists, and participants with no thyroid cancer diagnosis (N = 31). The HSDs went through several iterations over the course of a year, in collaboration with a highly engaged community advisory board, laying the groundwork for HSDs that are comprehensible, comparable, and appropriate for stated-preference research. FINDINGS: Thyroid cancer survivors compared their experiences with those described in the HSDs. Feedback included concern for the emotional well-being of study participants who would be reading them. Providers were attuned to the need for clinical accuracy and made suggestions to reflect their clinical experience, including for patients with complications or disease progression. The pilot participants with no thyroid cancer were particularly valuable in promoting the need to simplify language and maximize readability. DISCUSSION: Stakeholder engagement was critical to being responsive to feedback as the iterations were refined and presented. Continuous engagement and consultation with multiple sources strengthened the HSDs. A secondary outcome from this project is that stakeholders expressed interest in adapting the HSDs into decision aids for people newly diagnosed with low-risk thyroid cancer.

Treat Now or Treat Later: Comparative Effectiveness of Adjuvant Therapy in Resected Stage IIIA Melanoma.

BACKGROUND: Adjuvant therapy for most sentinel-node-positive (stage IIIA) melanoma may have limited clinical benefit for older patients given the competing risk of non-cancer death. The objective of this study is to model the clinical effect and cost of adjuvant therapy in stage IIIA melanoma across age groups. STUDY DESIGN: A Markov decision analysis model simulated the overall survival of patients with resected stage IIIA melanoma treated with adjuvant therapy vs observation. In the adjuvant approach, patients are modeled to receive adjuvant pembrolizumab (BRAF wild type) or dabrafenib/trametinib (BRAF mutant). In the observation approach, treatment is deferred until recurrence. Transition variables were derived from landmark randomized trials in adjuvant and salvage therapy. The model was analyzed for age groups spanning 40 to 89 years. The primary outcome was the number needed to treat (NNT) to prevent one melanoma-related death at 10 years. Cost per mortality avoided was estimated using Medicare reimbursement rates. RESULTS: Projections for NNT among BRAF wild type patients increased by age from 14.71 (age 40 to 44) to 142.86 (age 85 to 89), with patients in cohorts over the age of 75 having an NNT over 25. The cost per mortality avoided ranged from $2.75 million (M) (age 40 to 44) to $27.57M (age 85 to 89). Corresponding values for BRAF mutant patients were as follows: NNT 18.18 to 333.33; cost per mortality avoided ranged from $2.75M to $54.70M. CONCLUSION: Universal adjuvant therapy for stage IIIA melanoma is costly and provides limited clinical benefit in patients older than 75 years.

Low-Risk Papillary Thyroid Cancer: Treatment De-Escalation and Cost Implications.

INTRODUCTION: The American Thyroid Association (ATA) updated consensus guidelines in 2015 for radioactive iodine (RAI) and resection for low-risk papillary thyroid cancer. The objective of this study was to describe the evolution of institutional practice patterns and estimate the cost implications of these trends. MATERIALS AND METHODS: Patients with cT1-T2N0 papillary thyroid cancer were identified via an institutional tumor registry. Incidences of total thyroidectomy or RAI were tracked longitudinally using cumulative sum. Real-world costs for RAI and each surgical encounter were adjusted for inflation and standardized to national average costs from National Inpatient Sample cost data. RESULTS: Sixty-one patients met inclusion criteria between 2007 and 2018. Among these, 28 patients underwent total thyroidectomies and received RAI treatments based on criteria pre-dating the 2015 ATA guidelines. Cumulative sum revealed significant decreases in the rate of total thyroidectomy following May 2015 (15.8% versus 59.5%, P = 0.002) and RAI following March 2013 (3.0% versus 32.1%, P = 0.002). There were no locoregional recurrences in either period. The average cost savings attributable to these institutional practice changes was $1580 per patient. CONCLUSIONS: De-escalation in surgical and RAI utilization for low-risk papillary thyroid cancer according to 2015 ATA guidelines is associated with a substantial decrease in real-world costs.

Increasing utilization of endoscopic local excision for early-stage gastric cancers in the United States.

BACKGROUND: The Japanese Gastric Cancer Association provided updated criteria for endoscopic local excision of early-stage gastric cancer in 2018. The purpose of this study was to evaluate utilization patterns for endoscopic local excision in the United States for resectable gastric adenocarcinoma. METHODS: Patients with resectable gastric adenocarcinoma were identified from the National Cancer Database between 2010 and 2017. Patients were classified into strict appropriate criteria, expanded criteria, and inappropriate based on the Japanese Gastric Cancer Association guidelines. Factors associated with endoscopic local excision were identified using univariate and logistic multivariate regression. RESULTS: Within the National Cancer Database, 46,334 patients were stratified into strict appropriate criteria (n = 1,405), expanded criteria (n = 727), and inappropriate (n = 43,675). Annual cases of local excision increased by 76.9% over the study period, from 273 in 2010 to 483 in 2017. Among patients who underwent local excision, 10.1% were classified as strict appropriate criteria, 1.6% were classified as expanded criteria, and 84.5% were classified as inappropriate. Among inappropriate patients, factors associated with endoscopic local excision were: more recent year of diagnosis, increasing age, female sex, tumor located in the cardia, smaller size, low-grade, absence of lymphovascular invasion, and treatment at an academic facility. CONCLUSION: The use of endoscopic local excision for gastric cancer has nearly doubled since 2010. However, most patients do not satisfy consensus criteria for endoscopic therapy.

Extent of gastrectomy and lymphadenectomy for gastric adenocarcinoma.

Gastric adenocarcinoma is one of the most common and lethal cancers worldwide and is associated with a high frequency of nodal metastasis. The value of multimodality therapy is well-established, but gastric resection and locoregional lymph node dissection are important mainstays in potentially curative therapy. However, there has been considerable regional variation in surgical approach and debate regarding the ideal extent of gastric resection, gastric reconstruction, and extent of lymphadenectomy. This chapter outlines the current evidence in the surgical management of gastric adenocarcinoma. The advent of minimally invasive approaches to gastric operations is also discussed.

Enhanced Recovery After Major Gastrectomy for Cancer.

Enhanced recovery after surgery (ERAS) protocols have gained increasing popularity over the past 10 years, and its overarching objectives are to improve perioperative morbidity and reduce postoperative length of stay. Consensus guidelines from the ERAS Society specific to major gastrectomy were published in 2014, however since that time, prospective and retrospective studies have expanded the collective evidence for both the content and efficacy of ERAS pathways for gastrectomy. This objective of this review was to summarize recent data pertinent to the preoperative, perioperative, and postoperative management of gastrectomy patients along an ERAS pathway.