Liver resection for metastatic disease after y90 radioembolization: a case series with long-term follow-up.

INTRODUCTION: There are only few reports of liver resections for metastatic disease in patients previously treated with Y-90 radioembolization (RE), and long-term outcome data are sparse. We reviewed our center's experience in patients undergoing hepatectomy after hepatic RE. METHODS: A retrospective chart review of patients undergoing RE from 2004 to 2011 was performed. Demographic, clinicopathologic, operative, and long-term outcomes variables were collected. Independent pathologic review of tumor necrosis and normal liver tissue grading of fibrosis and inflammation after resection was performed. Data are expressed as medians and ranges. RESULTS: RE was delivered to 106 patients with primary and metastatic disease of the liver, of whom 9 patients (6 males, 3 females, median age 54 (47-76) years) with metastatic disease ultimately underwent resection. RE was previously administered to the right liver in five, the left liver in one, and to the whole liver in three. Two patients had a second RE performed before resection. Six of the nine patients had previously received several infusions of cytotoxic therapy. The operations occurred at a median of 115 (56-245) days after RE and included right lobectomy (n = 5), left lobectomy (n = 1), left-lateral sectionectomy (n = 1), and bilobar wedge resections (n = 2). Extrahepatic sites were resected in three patients. Median blood loss was 900 (range 250-3600) ml. Grade 3 or higher complications occurred in seven cases (78 %). Follow-up was complete all nine patients. Three patients (33 %) died within 30 days of resection. All those surviving the operative period had disease recurrence (time to recurrence: 202 [range 54-315] days), and all have since died (overall survival: 584 [range 127-1230] days). Review of resected specimens demonstrated median tumor necrosis of 70 % (range 20-90 %). In nontumor-bearing liver, fibrosis grade (0-4) and inflammation score (0-4) was 2 or less in all specimens. CONCLUSIONS: In this small cohort of highly selected and heavily pretreated patients, long-term survival in patients undergoing resection after RE appears possible, but the operations may carry substantial risks-highlighting the importance of careful patient selection for these resections. The etiology of morbidity and mortality is likely multifactorial and additional reports that include long-term outcomes will be necessary to identify more clearly the impact of RE on postoperative complications and death.

Technical effects of adding 1 % lidocaine to technetium sulfur colloid for sentinel lymphatic mapping in early breast cancer: analysis of data from a double-blind randomized controlled trial.

BACKGROUND: A practice standard in sentinel lymph node (SLN) mapping in breast cancer is intradermal injection of technetium-99m sulfur colloid (Tc-99m), resulting in significant patient discomfort and pain. A previous randomized controlled trial showed that adding lidocaine to Tc-99m significantly reduced radioisotope injection-related pain. We tested whether 1 % lidocaine admixed with Tc-99m affects feasibility of SLN mapping. METHODS: Between January 2006 and April 2009, 140 patients with early breast cancer were randomly assigned (1:1:1:1) to receive standard topical 4 % lidocaine cream and intradermal Tc-99m (control) or to one of three other study groups: topical placebo cream and injection of Tc-99m containing sodium bicarbonate (NaHCO3), 1 % lidocaine, or both. All SLN data were collected prospectively. RESULTS: Study groups were comparable for clinicopathological parameters. As previously reported, the addition of 1 % lidocaine to the radioisotope solution significantly improved patient comfort. Overall SLN identification rate in the trial was 93 %. Technical aspects of SLN biopsy were similar for all groups, including time from injection to operation, first SLN (SLN 1) gamma probe counts, ex vivo counts for SLN 1 and SLN 2, and axillary bed counts. SLN identification rates were comparable statistically: control (96 %), lidocaine (90 %), sodium bicarbonate (97 %), and sodium bicarbonate-lidocaine (90 %). The control group had a significantly higher SLN 2/SLN 1 ex vivo count ratio, and the number of SLNs detected was significantly reduced in the lidocaine versus no-lidocaine groups (p < 0.05). CONCLUSIONS: Addition of 1 % lidocaine to standard radioisotope solution for SLN mapping in breast cancer is associated with fewer SLNs detected, but it does not appear to compromise SLN identification.

National Economic Conditions May Impact the Financial Barriers to Travel for Cancer Operations.

Background: Better cancer-related outcomes are associated with physicians and hospitals with higher case volume. This serves as an incentive to refer patients requiring complex cancer operations to large referral centers, which may require increased travel for patients. However, barriers exist for patients to travel for cancer care, some of which may be aggravated or alleviated by factors relating to the health of the national economy. This impact may be reflected in variability of travel distances for cancer operations over time particularly for complex operation such as pancreatectomy and esophagectomy compared with less complex resections such as those for breast cancer or melanoma. Methods: We obtained the estimated travel distance for patients undergoing operations for cancer of the pancreas, esophagus, skin (melanoma), and breast from the National Cancer Database from 2004 to 2017 and correlated them with economic factors obtained from public sources. We then examined the impact of unemployment rates, gas prices, and inflation on travel distances regarding disadvantaged groups. Correlations were measured by the (rank-based, nonparametric) Spearman's correlation coefficient, and the corresponding P value is obtained by the asymptotic distribution of the coefficient. A P value of 0.05 equates to an absolute correlation value of 0.532. To adjust for multiple tests, a more restrictive P value of 0.01 was also assessed, which equates to correlation coefficients of absolute value greater than 0.661. Results: There were 4,222,380 cases in the dataset, of which 1,781,056 remained after exclusion. The economic factors that were associated most strongly with the distance patients traveled for all cancer operation types were the labor force participation rate, personal savings, consumer price index, and changes in gasoline prices. Inflation and rising gasoline prices were often inversely related with travel distance in lower-income and less well-educated regions and African American patients. Conclusions: Several macroeconomic factors correlate with the travel distance for operations, suggesting that the economic health of the nation may aggravate or alleviate the financial barriers to travel for cancer operations. Financially disadvantaged groups may be particularly vulnerable to changes in gasoline prices and inflation. Organizations serving these populations may need to increase patient support services during times of economic hardship to avoid the exacerbation of health care disparities.

The impact of recurrent laryngeal neuromonitoring on multi-dimensional voice outcomes following thyroid surgery.

BACKGROUND: Voice changes after thyroidectomy are common but not always related to recurrent laryngeal nerve (RLN) injury. We evaluated if RLN neuromonitoring correlated with non-RLN injury-related changes in voice after thyroidectomy. METHODS: Prospective multi-dimensional voice assessment was conducted on patients undergoing thyroidectomy before, 1-4 weeks, and 6 months postoperatively. Voice outcome (VO) was determined as normal (NormVO) or negative (NegVO) based upon combinations of patient-reported symptoms, videolaryngoscopy, a composite of acoustic measurements, and clinician-perceived voice quality. Groups with and without neuromonitoring were compared for early and durable differences in VO. RESULTS: Ninety-one patients underwent thyroidectomy; 39 with RLN neuromonitoring and 52 without. The two study groups were similar with regard to baseline characteristics including voice assessment. There was no difference in NegVO between neuromonitored and non-monitored patients at 1-4 weeks (n = 89; 32% vs. 27%; P = 0.81) and 6 months (n = 71, 14% vs. 7%; P = 0.42) after thyroidectomy. Neuromonitoring was associated with a 48-min increase in median operative time, but this finding was not statistically significant in a multivariate model. CONCLUSION: In this study, recurrent laryngeal neuromonitoring did not appear to influence non-RLN injury related VO as measured by a comprehensive multidimensional voice assessment.

Management of post-operative subcutaneous emphysema with the application of wound VAC therapy.

Extensive subcutaneous emphysema (SE) complicates between 1 and 6% of elective thoracic procedures. The management of SE is varied, and may include increasing the suction of chest tubes, placement of additional chest tubes, placement of subcutaneous drains and creation of releasing incisions. We present five patients with post-operative SE treated successfully with a subcutaneous infraclavicular incision and wound VAC therapy. A 5-cm incision was made 2 cm below the clavicle down and through the pectoralis major fascia. A VAC dressing was fitted to the wound and suction was applied to -125 mm Hg. Data were retrospectively collected and analyzed. VAC dressing was placed a median of 6 days after initial operation. All patients had improvement in symptoms and resolution of SE by VAC dressing therapy. Subcutaneous infraclavicular incision and VAC dressing placement is a viable treatment for patients with post-operative SE who fail conservative therapy.

Axillary sentinel lymph node biopsy after mastectomy: a case report.

BACKGROUND: Sentinel lymph node biopsy has been established as the preferred method for staging early breast cancer. A prior history of mastectomy is felt to be a contraindication. CASE PRESENTATION: A patient with recurrent breast cancer in her skin flap was discovered to have positive axillary sentinel nodes by sentinel lymph node biopsy five years after mastectomy for ductal carcinoma in situ. CONCLUSION: A prior history of mastectomy may not be an absolute contraindication to sentinel lymph node biopsy.

Standard versus pH-adjusted and lidocaine supplemented radiocolloid for patients undergoing sentinel-lymph-node mapping and biopsy for early breast cancer (PASSION-P trial): a double-blind, randomised controlled trial.

BACKGROUND: Sentinel-lymph-node (SLN) mapping and biopsy maintains staging accuracy in early breast cancer and identifies patients for selective lymphadenectomy. SLN mapping requires injection of technetium-99m-sulfur colloid-an effective but sometimes painful method, for which better pain-management strategies are needed. In this randomised, double-blind trial, we compared degree of pain between standard radiocolloid injection and pH-adjusted and lidocaine-supplemented formulations for patients undergoing SLN mapping for breast cancer. METHODS: Between Jan 13, 2006, and April 30, 2009, 140 patients with early breast cancer were randomly assigned in a 1:1:1:1 fashion to receive the standard topical 4% lidocaine cream and injection of [(99m)Tc]Tc-sulfur colloid (n=35), or to one of three other study groups: topical placebo cream and injection of Tc-sulfur colloid containing either sodium bicarbonate (n=35), 1% lidocaine (n=35), or sodium bicarbonate and 1% lidocaine (n=35). The randomisation sequence was computer generated, and all patients and investigators were masked to treatment allocation. The primary endpoint was patient-reported breast pain immediately after radioisotope injection, using the Wong-Baker FACES pain rating scale and McGill pain questionnaire, analysed in the per-protocol population. This study is registered with ClinicalTrials.gov, number NCT00940199. FINDINGS: 19 of the 140 patients enrolled were excluded from analysis: nine declined study participation or sought care elsewhere, nine did not undergo SLN mapping because of disease extent or a technical problem, and one had unreliable data. There were no adverse events. Mean pain scores on the Wong-Baker scale (0-10) were: 6.0 (SD 2.6) for those who received standard of practice, 4.7 (3.0) for those who received radiocolloid plus bicarbonate, 1.6 (1.4) for those who received radiocolloid plus 1% lidocaine, and 1.6 (1.3) for those who received radiocolloid plus bicarbonate and 1% lidocaine (p<0.0001). Mean pain rating, according to the McGill questionnaire (0-78), was 17.5 (SD 11.8) for the standard-of-care group, 15.4 (14.4) for the sodium bicarbonate group, 4.6 (4.5) for the 1% lidocaine group, and 3.4 (5.1) for the sodium bicarbonate plus 1% lidocaine group (p<0.0001). SLN identification rates for each group were: 96% for the standard of care, 97% for sodium bicarbonate, 90% for 1% lidocaine, and 90% for sodium bicarbonate plus 1% lidocaine group (p=0.56). INTERPRETATION: For centres that use radiocolloid injections for SLN mapping in patients with early breast cancer, the addition of 1% lidocaine to the radioisotope solution can improve patient comfort, without compromising SLN identification. FUNDING: US Military Cancer Institute, the Clinical Breast Care Project, and the Army Regional Anesthesia and Pain Management Initiative.

The influence of a gene expression profile on breast cancer decisions.

PURPOSE: The Oncotype Dx Recurrence Score (RS), is often employed in patients with estrogen receptor-positive, node negative (ER+LN-) breast cancer. We investigated the impact of the RS on actual chemotherapy administration and the effect of the assay on a panel of breast oncology experts. PATIENTS AND METHODS: The prospective adjuvant chemotherapy recommendations (prior to RS) and actual adjuvant therapy (after RS) for consecutive patients with ER+LN- breast cancer were recorded. After 6 months and with the same information, a panel of five experts made adjuvant therapy recommendations with and without RS and rated the strength of their recommendations. Rates of panel consensus, recommendation changes, and changes in recommendation strength were compared. RESULTS: There were 29 patients (28 women). RS results altered the plan for chemotherapy in 9 patients (31%); 7 of 13 patients (54%) initially recommended for chemotherapy did not receive it, and 2 of 16 (13%) received chemotherapy following initial recommendations against it. RS results changed the panel's chemotherapy recommendation in 7 patients (24%): 5 of 12 (42%) recommendations for changed to against, and 2 of 17 (12%) recommendations against changed to for chemotherapy. RS increased consensus by the panel 10%, but did not increase the reported strength in chemotherapy recommendations. CONCLUSIONS: RS results were associated with real-world decision changes in 31% of patients and 24% of panel recommendations and increased panel consensus by 10%. However RS did not increase the strength of panelist's recommendations.

Fat-containing lesions of the retroperitoneum: radiologic-pathologic correlation.

Retroperitoneal lesions represent a broad, diverse collection of entities; when they contain fat, the differential diagnosis, which ranges from benign to fully malignant lesions, substantially narrows. Lipomas rarely occur in the retroperitoneum; thus, fat-containing lesions in this location should never be dismissed as lipoma. Pelvic lipomatosis is the overgrowth of histologically normal fat in the extraabdominal compartments of the pelvis along the perirectal and perivesicular spaces. Infants and young children develop lipoblastomas rather than liposarcomas, which occur in older patients. Liposarcomas typically occur in patients 50-70 years old and manifest in multiple subtypes, with the most common being well-differentiated liposarcoma. Liposarcomas are histologically and radiologically protean, with no one imaging feature specific across the spectrum of subtypes. Hibernoma is a rare benign soft-tissue tumor composed of brown fat. Because hibernomas contain varied portions of brown and white fat as well as lesser amounts of myxoid material and spindle cells, their imaging features vary considerably. Teratomas are neoplasms that originate in pluripotent cells--benign or malignant germ cells--that give rise to a wide spectrum of mature or immature tissues that are foreign to the location in which they arise and which demonstrate varying amounts of organ formation. Myelolipoma, a benign tumor composed of mature fat and interspersed hematopoietic elements that resemble bone marrow, typically originate in an otherwise normal adrenal gland. Angiomyolipoma is composed of varying admixtures of blood vessels, smooth muscle cells, and adipose tissue; any one or two of these elements may predominate. Although these fat-containing lesions have overlapping radiologic features, use of demographic and clinical data helps refine the diagnostic options and treatment.

Development of a clinical decision model for thyroid nodules.

BACKGROUND: Thyroid nodules represent a common problem brought to medical attention. Four to seven percent of the United States adult population (10-18 million people) has a palpable thyroid nodule, however the majority (>95%) of thyroid nodules are benign. While, fine needle aspiration remains the most cost effective and accurate diagnostic tool for thyroid nodules in current practice, over 20% of patients undergoing FNA of a thyroid nodule have indeterminate cytology (follicular neoplasm) with associated malignancy risk prevalence of 20-30%. These patients require thyroid lobectomy/isthmusectomy purely for the purpose of attaining a definitive diagnosis. Given that the majority (70-80%) of these patients have benign surgical pathology, thyroidectomy in these patients is conducted principally with diagnostic intent. Clinical models predictive of malignancy risk are needed to support treatment decisions in patients with thyroid nodules in order to reduce morbidity associated with unnecessary diagnostic surgery. METHODS: Data were analyzed from a completed prospective cohort trial conducted over a 4-year period involving 216 patients with thyroid nodules undergoing ultrasound (US), electrical impedance scanning (EIS) and fine needle aspiration cytology (FNA) prior to thyroidectomy. A Bayesian model was designed to predict malignancy in thyroid nodules based on multivariate dependence relationships between independent covariates. Ten-fold cross-validation was performed to estimate classifier error wherein the data set was randomized into ten separate and unique train and test sets consisting of a training set (90% of records) and a test set (10% of records). A receiver-operating-characteristics (ROC) curve of these predictions and area under the curve (AUC) were calculated to determine model robustness for predicting malignancy in thyroid nodules. RESULTS: Thyroid nodule size, FNA cytology, US and EIS characteristics were highly predictive of malignancy. Cross validation of the model created with Bayesian Network Analysis effectively predicted malignancy [AUC = 0.88 (95%CI: 0.82-0.94)] in thyroid nodules. The positive and negative predictive values of the model are 83% (95%CI: 76%-91%) and 79% (95%CI: 72%-86%), respectively. CONCLUSION: An integrated predictive decision model using Bayesian inference incorporating readily obtainable thyroid nodule measures is clinically relevant, as it effectively predicts malignancy in thyroid nodules. This model warrants further validation testing in prospective clinical trials.

Physician derived versus administrative data in identifying surgical complications. Fact versus Fiction.

BACKGROUND: Administrative data are widely used as determinants of surgical quality. We compared surgical complications identified in a structured surgical review to coding and billing data of over a 19-month period. METHODS: A retrospective review of monthly morbidity and mortality conference reports was compared to a report over the same time period generated from hospital coding and billing data. RESULTS: 807 sequential operative procedures were included. Physician derived data compared to administrative data identified a complication of any severity in 205 (25.4%) versus 111 (13.8%) cases (r = 0.39), and major complications in 68 (8.4%) versus 46 (5.7%) cases (r = 0.36). Review of the administrative data regarding major complications identified 80 false negatives, 52 false positives, and 38 true positive designations. Overall sensitivity, specificity, positive and negative predictive values, and accuracy for administrative data in identifying major complications was 0.32, 0.99, 0.42, 0.99, and 0.99. CONCLUSIONS: The correlation between physician determined and administrative data with regard to identifying surgical complications is poor. Administrative data are insensitive and lack positive predictive value.

The functional impact on voice of sternothyroid muscle division during thyroidectomy.

BACKGROUND: Post-thyroidectomy voice dysfunction may occur in the absence of laryngeal nerve injury. Strap muscle division has been hypothesized as one potential contributor to dysphonia. METHODS: Vocal-function data, prospectively recorded before and after thyroidectomy from two high-volume referral institutions, were utilized. Patient-reported symptoms, laryngoscopic, acoustic, and aerodynamic parameters were recorded at 2 weeks and 3 months postoperatively. Patients with and without sternothyroid muscle division during surgery were compared for voice changes. Patients with laryngeal nerve injury, sternohyoid muscle division, arytenoid subluxation or no early postoperative follow-up evaluation were excluded. Differences between study groups and outcomes were compared using t-tests and rank-sum tests as appropriate. RESULTS: Of 84 patients included, 45 had sternothyroid division. Distribution of age, gender, extent of thyroidectomy, specimen size, and laryngeal nerve identification rates did not differ significantly between groups. There was a significant predilection for or against sternothyroid muscle division according to medical center. No significant difference in reported voice symptoms was observed between groups 2 weeks or 3 months after thyroidectomy. Likewise, acoustic and aerodynamic parameters did not differ significantly between groups at these postoperative study time points. CONCLUSION: Sternothyroid muscle division is occasionally employed during thyroidectomy to gain superior pedicle exposure. Division of this muscle does not appear to be associated with adverse functional voice outcome, and should be utilized at surgeon discretion during thyroidectomy.

Clinical implications of fibroblast activation protein in patients with colon cancer.

PURPOSE: Human fibroblast activation protein (FAP)/seprase is a 97-kDa surface glycoprotein expressed on tumor associated fibroblasts in the majority of epithelial cancers including colon adenocarcinomas. FAP overexpression in human tumor cells has been shown to promote tumor growth in animal models, and clinical trials targeting FAP enzymatic activity have been initiated. The primary objective of this study was to evaluate the clinical significance of stromal FAP in human colon cancers by immunohistochemisty. EXPERIMENTAL DESIGN: Sections of paraffin-embedded resected primary human colon cancer specimens from 1996 through 2001 within the Fox Chase Cancer Center tumor bank were stained with D8 antibody directed against FAP/seprase. Xenotransplanted human colorectal tumors in mice were examined similarly for stromal FAP in tumors of different sizes. Overall percentage of stromal FAP staining of the primary tumor was assessed semiquantitatively (0, 1+, 2+, 3+) and staining intensity was also graded (none, weak, intermediate, strong). Survival time and time to recurrence data were analyzed using Kaplan-Meier plots, log-rank tests, and Cox proportional hazards models. RESULTS: One hundred thirty-eight patients with resected specimens were available for study (mean follow-up, 1,050 days) with 6 (4%) stage I, 52 (38%) stage II, 43 (31%) stage III, and 37 (27%) stage IV patients. FAP was detected in >93% of specimens. Semiquantitative staining was scored as 1+ in 28 (20%), 2+ in 52 (38%), and 3+ in 49 (35%). FAP staining intensity was graded as weak in 45 (33%), intermediate in 48 (35%), and dark in 36 (26%). Stromal FAP was found to correlate inversely with tumor stage (semiquantitative, P = 0.01; intensity, P = 0.009) and with tumor size of the tumor xenograft model (correlation coefficient, -0.61; P = 0.047), suggesting that stromal FAP may have a greater role in the early development of tumors. Furthermore, greater stromal FAP for patients with known metastatic disease was associated with a decreased survival. CONCLUSION: Our data indicate that patients whose colon tumors have high levels of stromal FAP are more likely to have aggressive disease progression and potential development of metastases or recurrence. This study affirms the rationale for ongoing clinical investigations using FAP as a therapeutic target to disrupt FAP-driven tumor progression in patients with metastatic disease. It also suggests that the effects of FAP inhibition should be investigated in earlier-stage tumors, given its high levels and potential effect earlier in the course of the disease.

Quality measurement affecting surgical practice: Utility versus utopia.

The Triple Aim: improving healthcare quality, cost and patient experience has resulted in massive healthcare "quality" measurement. For many surgeons the origins, intent and strengths of this measurement barrage seems nebulous-though their shortcomings are noticeable. This article reviews the major organizations and programs (namely the Centers for Medicare and Medicaid Services) driving the somewhat burdensome healthcare quality climate. The success of this top-down approach is mixed, and far from convincing. We contend that the current programs disproportionately reflect the definitions of quality from (and the interests of) the national payer perspective; rather than a more balanced representation of all stakeholders interests-most importantly, patients' beneficence. The result is an environment more like performance management than one of valid quality assessment. Suggestions for a more meaningful construction of surgical quality measurement are offered, as well as a strategy to describe surgical quality from all of the stakeholders' perspectives. Our hope is to entice surgeons to engage in institution level quality improvement initiatives that promise utility and are less utopian than what is currently present.

Glenohumeral joint ganglion cyst and other rare cysts of the axilla.

Although ganglion cysts have been reported to arise from almost any joint, those arising from the glenohumeral joint producing an axillary mass are extremely rare. We report what we believe to be the eighth such case and describe its management. The unusual differential diagnosis and aids to diagnosis of axillary cysts are reviewed. Specific issues regarding axillary space ganglions are emphasized.

Methemoglobinemia: early intraoperative detection by clinical observation.

OBJECTIVES/HYPOTHESIS: Head and neck surgery patients often receive topical anesthetics in the office for examination or in the operating room for nasotracheal intubation. Benzocaine (a component of many topical preparations) is frequently employed for this purpose. Acute methemoglobinemia is a rare but potentially lethal complication of benzocaine administration. Early recognition and treatment may prevent complications or death. We report a case of acute intraoperative methemoglobinemia, caused by benzocaine spray, which was diagnosed (and reversed) quickly after the clinical observation of "chocolate brown blood" in the surgical field.

Voice outcomes after total thyroidectomy, partial thyroidectomy, or non-neck surgery using a prospective multifactorial assessment.

BACKGROUND: Voice alteration remains a significant complication of thyroid surgery. We present a comparison of voice outcomes between total thyroidectomy (TT), partial thyroidectomy (PT), and non-neck (NN) surgery using a multifactorial voice-outcomes classification tool. STUDY DESIGN: Patients with normal voice (n = 112) were enrolled between July 2004 and March 2009. The patients underwent TT (n = 54), PT (n = 35), or NN (n = 23) surgery under general endotracheal anesthesia as part of a prospective observational study involving serial multimodality voice evaluation preoperatively, and at 2 weeks, 3 months, and 6 months postoperatively. Patients with adverse voice outcomes were grouped into the negative voice outcomes (NegVO) category, including patients with objective (abnormality on videolaryngostroboscopy and substantial voice dysfunction) and subjective (normal videolaryngostroboscopy but with notable voice impairment) NegVO. Voice outcomes were compared among study groups. RESULTS: Negative voice outcomes occurred in 46% (95% CI, 34-59%) and 14% (95% CI, 6-30%) of TT and PT groups, respectively. No NegVOs were observed after NN surgery. Early NegVOs were more common in the TT group than in the NN or PT groups (p < 0.001). Most voice disturbances resolved by 6 months (TT 84%; PT 92%) with no difference in NegVO among all groups (p = 0.23). Black race and significant changes in certain voice outcomes measures at the 2-week follow-up visit were identified as predictors of late (3 to 6 months) NegVO. CONCLUSIONS: This comprehensive voice outcomes study revealed that the extent of thyroidectomy impacts voice outcomes in the early postoperative period, and identified risk factors for late NegVO in post-thyroidectomy patients who should be considered for early voice rehabilitation referral.