Yale UniversityEliScholar – A Digital Platform for Scholarly Publishing at Yale

Yale Medicine Thesis Digital Library School of Medicine

January 2016

Short-Term Outcomes And Long-Term PrognosisIn Oral Cavity CancerZachary Gregg SchwamYale University, zachary.schwam@yale.edu

Follow this and additional works at: http://elischolar.library.yale.edu/ymtdl

This Open Access Thesis is brought to you for free and open access by the School of Medicine at EliScholar – A Digital Platform for ScholarlyPublishing at Yale. It has been accepted for inclusion in Yale Medicine Thesis Digital Library by an authorized administrator of EliScholar – A DigitalPlatform for Scholarly Publishing at Yale. For more information, please contact elischolar@yale.edu.

Recommended CitationSchwam, Zachary Gregg, "Short-Term Outcomes And Long-Term Prognosis In Oral Cavity Cancer" (2016). Yale Medicine ThesisDigital Library. 2079.http://elischolar.library.yale.edu/ymtdl/2079

1

SHORT-TERM OUTCOMES AND LONG-TERM PROGNOSIS

OF ORAL CAVITY CANCER

A Thesis Submitted to the

Yale University School of Medicine

in Partial Fulfillment of the Requirements for the

Degree of Doctor of Medicine

by

Zachary G. Schwam and Benjamin L. Judson (advisor)1

1Department of Surgery, Section of Otolaryngology, Yale University School

of Medicine, New Haven, CT

2016

2

SHORT-TERM OUTCOMES AND LONG-TERM PROGNOSIS OF ORAL CAVITY

CANCER. Zachary G. Schwam, Benjamin L. Judson (advisor). Section of

Otolaryngology, Department of Surgery, Yale University School of Medicine, New

Haven, CT.

We sought to characterize short-term morbidity and mortality outcomes as well as

long-term changes in prognosis for oral cancer patients. We predicted that clinical and

demographic variables would affect short- and long-term outcomes. Retrospective

analyses of the National Surgical Quality Improvement Program (NSQIP) and National

Cancer Database (NCDB) were performed on 408 and 13,655 patients, respectively. Chi-

square, Kaplan Meier, logistic regression, and Cox proportional hazards regression were

performed. In the NSQIP, the overall adverse event and mortality rates were 20.3% and

1.0%, respectively. The most common complications were reoperation, infection, and

respiratory complications. Over 90% of post-discharge complications occurred by post-

discharge day (PDD) 14, but the majority of surgical-site infections and dehiscences

occurred by PDD 7. Neck dissection, smoking, and weight loss were associated with

several complications in multivariate analysis. In the NCDB, three-year overall survival

increased by 36.2% and 16.0% for patients with early and late stage disease (LSD),

respectively. In LSD patients, adjuvant chemoradiotherapy increased from 8.3% to 36.4%.

Later year of diagnosis (hazard ratio [HR] 0.76), neck dissection (HR 0.90), and negative

margins (HR 1.00) were associated with better overall survival (all p≤.002). Many

patients with oral cavity cancers experience postoperative complications, some of which

occur post-discharge. Early follow-up should be sought for high-risk patients. Long-term

prognosis for oral cancers has increased dramatically, as has the administration of

adjuvant chemoradiotherapy in LSD. Numerous sociodemographic, clinical, and

treatment variables may account for this difference in survival.

3

Acknowledgements:

I would like to thank Dr. Benjamin L. Judson for his mentorship and guidance,

the department of surgery for their generous financial support, and the section of

otolaryngology for demonstrating the interplay of research, education, and clinical

medicine at the highest levels. Additionally, the members of the office of student research,

namely Mae Geter and Donna Carranzo, have been instrumental in securing funding and

ensuring that my research needs have been met throughout the years. I would also like to

thank my family and friends for their support and understanding throughout my academic

career. These projects were funded by the NCATS CTSA-TL1 and James D. Hirsch

Medical Student Research Fellowships. Two published manuscripts serve as the basis for

this work, with permissions to reproduce text and images secured by the publishers:

Schwam ZG, Sosa JA, Roman SA, Judson BL. Complications and mortality

following surgery for oral cavity cancer. Laryngoscope 2015;125:1869-73 (1).

Schwam ZG, Judson BL. Improved prognosis for patients with oral cavity

squamous cell carcinoma: Analysis of the National Cancer Database 1998-2006.

Oral Oncol 2016;52:45-51 (2).

4

Table of Contents

Introduction………………………………………………………………………...p.1-11

Purpose and Aims…………………………………….…………………………...p.11-12

Methods…………...…………………………………………………………….…p.12-18

Results……………………….…………………………….……………………….p.18-42

Discussion……………………………………………………………………….....p.42-52

Conclusions……………………………………….…………………………….....p.52-53

References…………………………………………………………………………p.54-77

1

Introduction

Sub-sites and lymphatic drainage patterns of the oral cavity

The oral cavity is composed of seven sub-sites, namely the lip, oral tongue,

alveolar ridge, retromolar trigone, floor of mouth, buccal mucosa, and hard palate (3).

While the anatomy and lymphatic drainage of the oral cavity is varied and complex, some

patterns have emerged in terms of sites of nodal metastases from oral cancers. In a

review of 1,081 patients at Memorial Sloan Kettering Cancer Center undergoing radical

neck dissections between 1965 and 1986, 562 neck dissections were performed for oral

cavity malignancies (4). In those patients, lymph node metastases were found to

predominate in cervical lymph node levels I, II, and III, and were rarely found in levels

IV or V. “Skip lesions” were also not observed, in that any patients with level V nodal

metastases also had metastases in levels I-IV.

Epidemiology of oral cancer

It is estimated that 59,340 new primary cancers of the head and neck were

diagnosed in the United States in 2015, with 12,290 associated deaths (5). Primary

tumors of the oral cavity were estimated to cause over 30,000 of those cases (51%) and

nearly 6,000 (49%) deaths. Oral cavity cancer is also common worldwide, and is the

fifteenth most common malignancy according to the World Health Organization (6). Oral

cancers are more common in men, which is thought to be secondary to higher rates of

tobacco and alcohol use, as well as greater exposure to sunlight via outdoor occupations

in the case of lip cancer (7). Oral cancer is predominantly a disease of middle age, with

only 6% of cases reported in patients under 45 years of age (8). The median age at

2

diagnosis in the United States was 62.0 years between years 2000 and 2004 (6,9),

although there are reports of an increasing incidence of oral tongue squamous cell

carcinoma in young white males aged 20-44 years (10).

Risk factors for developing oral cavity squamous cell carcinoma

Risk factors for the development of oral cavity cancer have been well-

characterized, and overlap with those of many of the other head and neck sub-sites.

Tobacco and alcohol use have been associated with the development of many head and

neck cancers, and have been estimated to cause nearly three-quarters of oral cavity cancer

cases (11). Tobacco is found in many forms, with cultural differences in its consumption

related to distinct anatomic locations of lesions within the oral cavity. Dipping snuff is

common in the Southern United States, and has been associated with a 50-fold higher

incidence of buccal and gingival lesions (12). Similarly, the large percentage of the

population in Bombay, India that chews pan, a tobacco derivative, contributes to very

high rates of buccal cancers (13). Additionally, hard palate carcinomas are particularly

common amongst Indian women who engage in “reverse chutta smoking,” in which the

lit end of a cigar is placed inside the mouth (14). Smoking cigarettes, pipes, and cigars

have also linked to the development of oral cavity cancers, with pipe and cigar smokers at

higher risk of developing oral cancers than cigarette smokers (15). Tobacco products all

exhibit a dose-response tumorigenic effect (16), with polycyclic aromatic hydrocarbons

(PAHs), nicotine-derived nitrosamine ketones (NNKs), and N-nitrosonornicotine (NNN)

creating DNA adducts in addition to oxidative damage by nitric oxide and quinones (17).

Tobacco use is not only linked to development of head and neck squamous cell

3

carcinomas, but is also associated with more aggressive disease; patients with a history of

tobacco use are more likely to have regional metastases and extracapsular spread when

compared to those without such a history (18).

Worldwide, 3.6% of cancers are related to alcohol consumption, notably the oral

cavity, pharynx, larynx, liver, breast, esophagus, and colon (19). Ethanol consumption

has been observed to have a dose-response effect, with a progressively higher relative

risk of developing oral cancer with increasing ethanol intake. In certain case-control

studies, drinking ≥6 whiskey-equivalents daily had a higher relative risk than did

smoking ≥40 cigarettes daily on the development of oral carcinomas (20). While the

exact mechanism behind ethanol’s association is as of yet unknown, it is thought to

behave synergistically with tobacco through a direct topical effect (13,21). Mechanisms

proposed include mucosal exposure to high concentrations of acetaldehyde, a breakdown

product of ethanol generated by oral bacteria (19,22) that interferes with DNA synthesis

and repair by binding to enzymes involved in cytosine methylation and glutathione

production (19,23,24). Acetaldehyde also binds to DNA, forming mutagenic adducts

(25). Oral acetaldehyde concentrations have been found to be higher in smokers, with the

theory that smoking somehow shifts the native flora to more acetaldehyde-producing

species (26). Cigarette smoke itself has also been found to have rather high

concentrations of acetaldehyde.

Cancers of the lip have a distinct epidemiological pattern when compared to those

of the oral cavity, with additional risk factors including exposure to sunlight, pipe

4

smoking, fair complexion (27), and immunosuppression (28). Pipe smoking is thought to

cause lip cancer through multiple mechanisms, including mechanical irritation, thermal

injury, and chemical exposure (3,29). In Australia, Canada, and Spain, lip cancer is the

most common sub-site of the oral cavity (30). It is important to distinguish cancers of the

lip from surrounding sub-sites, notably the buccal mucosa and adjacent skin.

Additional risk factors for cancer of the oral cavity include poor dental hygiene,

(specifically not brushing one’s teeth every day) and wearing ill-fitting dentures that have

caused sores (31). While the effects of poor dentition can be measured, they tend to be

greatly outweighed by the effects of tobacco and alcohol (32).

Histology common to oral cancers

In an analysis of the National Cancer Database (NCDB) examining 58,976

patients with oral cancer, the most common subtypes were found to be squamous cell

carcinoma (86.3%), adenocarcinoma (5.6%), and verrucous carcinoma (2.0%). The

remaining were split evenly by carcinoma not otherwise specified, lymphoma, Kaposi

sarcoma, and “other” (33).

While conventional squamous cell carcinoma is the most common subtype,

multiple variants can be found within the oral cavity, including sarcomatoid, basaloid,

and verrucous squamous cell carcinoma (3). Sarcomatoid squamous cell carcinoma stains

positively for keratin, has squamous cells intermingling with spindle-type cells, and has a

similar overall survival to that of conventional squamous cell carcinoma. There is,

5

however, a higher rate of locoregional recurrence (34). Basaloid squamous cell

carcinomas typically have basaloid cells, prominent peripheral nuclear palisading (35),

and have a predilection for the oral tongue (36). Basaloid lesions tend to have similar

survival curves to moderate to poorly differentiated squamous cell carcinomas (36). In

verrucous lesions, which are commonly found on the buccal mucosa, there is typically a

thick zone of non-proliferating and non-keratinizing cells. It is considered low-grade in

nature and resistant to radiation (3). Numerous additional pathologies of the oral cavity

may be found, including minor salivary gland tumors, melanoma, sarcomas (including

Kaposi’s sarcoma), lymphoma, and granulomas.

Treatment options and factors affecting treatment

The treatment of oral cavity (including lip) cancers is guided by clinical and

pathological staging information, presence of high-risk pathologic features, and success

of surgical resection. Surgical resection of the primary is the backbone of treatment, with

elective or therapeutic neck dissection based on presence of clinically positive nodes,

tumor thickness, or the results of sentinel lymph node biopsy. Adjuvant radiotherapy can

be considered in the case of having one positive node without having adverse features,

but is indicated and frequently combined with chemotherapy if there is extracapsular

extension, positive margins, or other adverse features such as pathologic T3/T4

classification, N2/N3 nodal disease, nodal disease in levels IV or V, as well as perineural

and lymphovascular invasion. Definitive radiotherapy is also an option in the case of

small lesions without nodal metastases, with salvage surgery reserved for recurrence (37).

6

Presence of nodal metastasis is an important prognostic factor in oral cancers; as a

result, there has been much effort to determine predictors of occult nodal metastases in

the clinically negative neck (38,39), with rates of occult nodal metastases in T1/T2

lesions as high as 27-40% (40,41,42). Tumor thickness is thought to reflect proximity to

lymphatic and vascular structures (43) and to approximate aggressiveness of tumor

growth (44). It has also been postulated that tumor emboli are more apt to spread along

the wider lymphatic channels located in the deeper aspects of the tissue (45). Tumor

thickness ≥4mm is considered to be an indication for elective neck dissection, as 4.5% of

patients are expected to have occult metastases with a thickness of 4mm, while 16.6%

have them with a cutoff of 5mm (46). A recent randomized control trial of node-negative

oral cancer patients demonstrated a superior overall and disease-free survival for patients

undergoing elective as compared to therapeutic neck dissection (47).

Reconstruction after oncologic resection

For small lesions of the oral cavity, primary closure, placement of a split-

thickness skin graft, or healing by secondary intention may be considered. Larger lesions

may require free-tissue transfer or a regional pedicle flap, and are dependent on surgeon

preference and ability as well as patient expectations (3).

Effects of disease and treatment on quality of life

The oral cavity is an integral anatomic region, playing important roles in speech,

deglutition, taste, salivation, and mastication. As a result, tumor-induced dysfunction as

well as treatment-related effects may have a profound impact on a patient’s day-to-day

7

functioning. Numerous factors have been found to affect quality of life in oral cavity

cancer patients, including the site of tumor, tumor size, depressive symptoms, pre-

operative performance status, method of reconstruction, and whether a patient underwent

neck dissection (48). While there have been articles reporting no effect of tumor site on

quality of life, it has also been demonstrated that more posterior tumors may have a

profound effect based on the impact on deglutition (49). Similarly, larger tumors and

overall stage are often associated with worse pain and physical symptoms, especially

when treated with multimodality therapy (48, 50). Anxiety and depression are prevalent

among those with oral cancers (51), and disfigurement often leads to changes in self-

perception, isolation, and problems with partners (52).

Much attention has been paid to the effect of surgical reconstruction on quality of

life, with some authors reporting that primary closure and laser surgery have superior

outcomes to free-flap reconstruction one year post-operatively (48,49). In a prospective

trial examining quality of life after free tissue transfer for oropharyngeal cancer

reconstruction, an initial decline in performance was observed, but recovered by 6

months. Mental health and emotional indices were superior to pre-treatment levels at one

year post-operatively (53). No differences in quality of life measures have been noted in

post-maxillectomy patients undergoing reconstruction with a free flap versus obturator

(54), or for bony free flap versus marginal mandibulectomy in resection of large

mandibular tumors (55). Extent of neck dissection has also been related to postoperative

shoulder dysfunction and pain, most notably in patients undergoing bilateral neck

dissections of levels I-IV or any dissection including level V (56).

8

Synchronous and metachronous tumors

Patients with head and neck squamous cell carcinomas have or develop a second

primary in 10-40% of cases, with common sites including the head and neck, esophagus,

and lung (57). Synchronous lesions are those that are diagnosed within 6 months of the

initial cancer, and metachronous lesions occur after the 6-month window has passed.

Panendoscopy of the upper aerodigestive tract and the tracheobronchial tree at the time of

diagnosis routinely yields a second primary tumor in 9-14% of cases (58,59). Up to 33%

of patients with oral cavity primaries have been found to have second primaries, with

most being found in the oral cavity, pharynx, larynx, lung, and esophagus (58). The risk

of developing a second oral cancer is highest in patients with primaries of the gingiva,

floor of mouth, and buccal mucosa, with gingival and floor of mouth primaries most

likely to have their second lesions in the lung (60,61,62). Identification of a second

primary was linked to a substantial decrease in five-year overall survival from 26.9% to

15.2% in one series (63).

Role of adjuvant therapy in treating oral cancer

Before 2004, locoregionally advanced head and neck squamous cell carcinoma

was typically treated with surgical resection and adjuvant radiotherapy. With this

treatment algorithm, rates of locoregional recurrence, distant metastasis, and five-year

overall survival rates were 30%, 25%, and 40%, respectively (64,65). The Radiation

Therapy Oncology Group (RTOG) 9501 and European Organization for Research and

Treatment of Cancer (EORTC) 22931 trials, which were both published in 2004, showed

9

a significant improvement in disease-free survival and locoregional control for high-risk

and late stage head and neck cancer patients. The two trials, however, had discordant

results with respect to overall survival, with EORTC 22931 showing a significant

improvement, and RTOG 9501 only demonstrating minimal improvement (2,66,67).

While different end-outcomes and definitions of “high risk” were used, a combined

follow-up analysis demonstrated the greatest benefit of adjuvant chemoradiotherapy to be

in those with positive margins or ECE (2,65).

Effect of screening in oral cancer

Currently, the United States Preventive Services Task Force maintains that there

is insufficient evidence for routine oral cancer screening in asymptomatic adults (68).

While routine oral cancer screening has not gained traction in the United States, there has

been some success on an international level. In a longitudinal screening program in

Kerala, India, nearly 200,000 people were randomized to either undergo serial visual

screenings or be placed in a control group over a thirteen-year period. While a reduction

of 12% in oral cancer mortality did not reach significance when all patients were

combined, tobacco/alcohol users in the serial screening group had a 38% lower incidence

of oral cancer and 81% lower oral cancer mortality when compared to patients in the

control group (69). Other efforts have focused on salivary biomarker detection in high-

risk patients (70), as well as autofluorescence visualization devices, cytological brush

tests (71), and differential gene expression using microarrays (72).

Current literature on short term outcomes and long term prognosis

10

The five-year relative survival rates for many head and neck cancers have

improved over the years, with increases from 53% (1975-77) to 66% (2004-10) for oral

cavity and pharyngeal primaries (5). This is in contrast to laryngeal primaries, which

have seen a significant decrease in their five-year relative survival from 66% to 63% over

the same period. This may be due to roughly two-thirds of head and neck cancer patients

presenting with advanced disease (73), and between 4-25% eventually developing distant

metastases. It has been found that between 7% and 12% have distant metastases at

presentation (73,74,75). Even in light of advances in diagnostic imaging and methods

and increased awareness, the proportion of head and neck cancer patients presenting with

Stage IV disease has not changed appreciably since 1990 (76).

Data on changes in prognosis of oral cavity malignancies are varied, with some

reports stating that the five-year survival rate has remained relatively stable over several

decades at 50-55%, with variations based on several variables including race and gender

(77,78,79). Oral cavity sub-site has also been shown to have prognostic value, with

cancers of the oral tongue showing a worse overall and cause-specific survival when

compared to other oral sub-sites (80). The long-term prognosis for lip cancer, however,

is excellent, with five-year survival rates of 95% (77,79). These data are in contrast to an

international cohort study from seven leading cancer centers that demonstrated a

significant improvement in survival for oral cancer patients (81), and a report from

Memorial Sloan Kettering Cancer Center showing a significant increase in survival for

oral tongue cancers between 1978 and 1987 (82).

11

Unfortunately, postoperative complications are rather common in head and neck

cancer patients, with reported complication rates as high as 53% (83,84,85,86). Wound

infections at the surgical site are the most common (85) in the literature, with risk factors

including neoadjuvant therapy (86), advanced disease (83,84,85), comorbidity, and neck

dissection (83). In a retrospective review of 1,693 Taiwanese oral cavity cancer patients

undergoing primary resection, the wound infection rate was found to be 19.8%, with risk

factors including diabetes, flap reconstruction, and low postoperative serum albumin

level (83). Thirty-day readmission and mortality has been reported using the NCDB for

patients with oral cavity squamous cell carcinoma (87), and Chen et al. have examined

postdischarge complications for all otolaryngology procedures in the NSQIP (88). As

postdischarge complications are one of the most important contributors to readmission

(89) and may delay the receipt of critical adjuvant therapies, investigating them remains

critical.

Purpose, aims, and hypotheses

The purpose of this study was multifold, and included characterizing both short-

term morbidity and mortality outcomes as well as long-term changes in prognosis for

patients with cancers of the oral cavity using large administrative datasets. We

hypothesized that clinical and demographic variables such as age and various

comorbidities would affect short-term postoperative morbidity and mortality, and sought

to characterize modifiable risk factors to predict which patient populations were at

highest risk. It was also our aim to characterize the time course of postoperative and post-

discharge complications (1). While we anticipated the majority of complications to occur

12

during admission, we believed that certain adverse events would tend to occur after

discharge. In identifying risk factors for postoperative adverse events, we predicted that

more preoperative comorbidity would be re associated with more morbidity and mortality

postoperatively. In investigating the long-term prognosis for patients with oral cavity

squamous cell carcinoma, we predicted that three-year overall survival would increase

over the study period, and would likely be associated with the administration of adjuvant

therapies, superior surgical techniques as measured by rates of negative margins, as well

as socio-demographic and oncologic variables such as age, cancer stage, and the category

of treatment center (2).

Methods

Division of responsibility

All procedures, methods, and statistical analyses were performed principally by

the first author. This included selecting the cohort, choosing inclusion and exclusion

criteria, recoding and redefining variables, univariate and multivariate analysis, creation

of tables and figures, and authorship of the manuscript. Data interpretation was chiefly

performed by the first author with assistance from the faculty advisor.

Description of the National Surgical Quality Improvement Program§

To characterize and determine the timing of postoperative complications

following oncologic resection of oral cavity malignancies, we utilized the American

§Adapted with permissions from: Schwam ZG, Sosa JA, Roman SA, Judson BL. Complications and mortality

following surgery for oral cavity cancer: analysis of 408 cases. Laryngoscope 2015;125:1869-73.

13

College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP)

Participant Use Data File (PUF) to examine data from patients undergoing surgery

between 2005 and 2010 (1). The NSQIP collects data from over 400 participating sites,

and is prospective, risk-adjusted, and validated against claims data in recording major

postoperative complications (1,90,91). The NSQIP is the only national dataset of its kind.

Data on 135 variables detailing preoperative patient characteristics and comorbid

conditions, intraoperative processes, and postoperative morbidity and mortality are

carefully recorded by a Surgical Clinical Nurse Reviewer at each site, and data are not

limited to in-hospital events (1,92,93,94). Due to a robust auditing and training process,

the inter-rater reliability of variables in the database has improved over the years,

ensuring standardization of data inputs (95). The NSQIP clinical reviewers rely on an

eight-day cycle in which representative cases are selected from the operative logs to

ensure lack of bias in selection; using this process, cases “have an equal chance of being

selected from each day of the week (96).”

Description of the National Cancer Database†

The NCDB was used to determine changes in long-term prognosis of patients

with oral cavity squamous cell carcinoma diagnosed between years 1998-2006. The

NCDB is a joint venture of the American Cancer Society and the Commission on Cancer

(CoC) of the American College of Surgeons, and has been in existence since 1989 (2,97).

It accounts for more than 1440 CoC-approved cancer program registries (2,76,98),

approximately 75% of cancers diagnosed in the United States, and is the pre-eminent

†Adapted with permissions from Schwam ZG, Judson BL. Improved prognosis for patients with oral cavity squamous

cell carcinoma: analysis of the National Cancer Database 1998-2006. Oral Oncol 2016;52:45-51.

14

oncologic database in the United States. The NCDB uses the same definitions and

standards as federal cancer registry systems (99), with coding guidelines found in the

Registry Operations and Data Standards (100) and Facility Oncology Registry Data

Standards manuals (2,101). Data is collected on socio-demographic, clinical, oncologic,

and treatment-related variables.

Selection criteria using the ACS NSQIP§

The NSQIP was queried by International Classification of Diseases-9th edition

(ICD-9) codes (102), selecting for malignant lesions of the oral cavity: 141.1-141.5,

144.0, 144.1, 144.8, 144.9, 145.0-145.2, 145.6, 145.8, and 145.9 (1). The following

Current Procedural Terminology (CPT) codes were then included in the analysis either as

the principal or concurrent procedure: 21025, 21026, 21034, 21040, 21044, 21045, 21050,

21198, 21299 (bony excision), 21557 (radical tumor resection), 31225 (maxillectomy),

41116 (excision from floor of mouth), 40810, 40812, 40814, 40816, 40899 (excision

from vestibule of mouth), 41110, 41112, 41114, 41120, 41130, 41135, 41140, 41145,

41599 (excision of tongue lesion, glossectomy with or without neck dissection), 41150,

41153, 41155 (glossectomy as part of composite procedure, with or without neck

dissection), 41825, 41827, 41830 (excision of dento-alveolar structures), 42104, 42106,

42107, 42120, 42140, 42160, 42299 (excision or destruction palatal or uvular lesion),

42415 (excision of parotid tumor), 42842, 42844, and 42845 (radical resection of

retromolar trigone with or without flap reconstruction) (1,103).

§Adapted with permissions from: Schwam ZG, Sosa JA, Roman SA, Judson BL. Complications and mortality

following surgery for oral cavity cancer: analysis of 408 cases. Laryngoscope 2015;125:1869-73.

15

Preoperative and intraoperative variables in the NSQIP§

Definitions of variables included in the database are described in the ACS-NSQIP

user guide (104) with several modifications (1). “Pulmonary comorbidity” was defined

as having dyspnea on exertion or at rest, a history of chronic obstructive pulmonary

disease, or being ventilator-dependent. “Cardiovascular comorbidity” was defined as a

history of congestive heart failure, angina, myocardial infarction (MI), peripheral

vascular disease, percutaneous coronary intervention, or cardiac surgery. “Poor

functional status” prior to surgery was defined as being either partially or totally

dependent on others for activities of daily living. Neck dissection was determined by the

following CPT codes (1,103): 38542 (dissection of deep jugular nodes), 38700, 41153

(suprahyoid neck dissection), 38720, 38724, 41135, 41155, and 42426 (includes

complete lymphadenectomy, modified radical neck dissection, radical neck dissection, or

neck dissection in addition to another procedure). Cases in which neck dissection was the

only procedure performed were excluded from analysis (1).

Postoperative outcomes in the NSQIP§

Complications and adverse events analyzed included surgical-site (superficial,

deep, and organ-space, and wound-dehiscence), infectious (urinary tract infection,

pneumonia, sepsis, and septic shock), respiratory (re-intubation, failure to wean, or

ventilator use >48 hours), renal (acute renal insufficiency or acute renal failure), and

cardiovascular complications (myocardial infarction, cardiac arrest, or stroke), as well as

flap failure, venous thromboembolism (deep venous thrombosis or pulmonary embolism),

§Adapted with permissions from: Schwam ZG, Sosa JA, Roman SA, Judson BL. Complications and mortality

following surgery for oral cavity cancer: analysis of 408 cases. Laryngoscope 2015;125:1869-73.

16

reoperation, and death (1,105).

Selection Criteria using the NCDB†

We extracted oral cancer patient data from the NCDB using the following

International Classification of Diseases for Oncology, 3rd edition (ICD-O-3) (2,106)

topography codes: C00.0, 00.1, 00.2, 00.3, 00.4, 00.5, 00.6, 00.8, 00.9 (Lip), 02.0, 02.1,

02.2, 02.3, 02.8, 02.9 (oral tongue), 03.0, 03.1, 03.9 (gingiva and alveolus), 04.0, 04.1,

04.8, 04.9 (floor of mouth), 05.0, 05.8, 05.9 (hard palate), 06.0, 06.1 (buccal mucosa),

06.2 (retromolar trigone), 06.8, and 06.9 (other and unspecified mouth). We then

selected for patients with squamous cell carcinomas using the following ICD-O-3

morphology codes 8051, 8052, 8070-8076, 8078, 8083, and 8084 (2,105). Patients with

T0, TX, Tis, and NX lesions as well as those with Stage 0 disease were excluded.

Staging information reflects the American Joint Committee on Cancer (AJCC) staging

system, 5th (1998-2002) and 6th editions (2003-2006) (107,108). Patients with incomplete

information for the following variables of interest were not included in our analyses:

patient gender, age, race, Hispanic origin, insurance status, facility type, TNM

classifications and overall stage, information on first course of treatment (surgery,

chemotherapy, radiation, surgical margins, scope of regional lymph node surgery), last

contact/death, and vital status at follow-up (2).

Variable definitions in the NCDB†

†Adapted with permissions from Schwam ZG, Judson BL. Improved prognosis for patients with oral cavity squamous

cell carcinoma: analysis of the National Cancer Database 1998-2006. Oral Oncol 2016;52:45-51.

†Adapted with permissions from Schwam ZG, Judson BL. Improved prognosis for patients with oral cavity squamous

cell carcinoma: analysis of the National Cancer Database 1998-2006. Oral Oncol 2016;52:45-51.

17

Variable definitions are as defined in the NCDB data dictionary (109), with some

exceptions. Patients were coded as diagnosed in years 1998-2003 or 2004-2006, as the

landmark RTOG 9501 and EORTC 22931 trials involving the use of adjuvant therapy in

head and neck cancer were first published in 2004 (2,65,66,67). Hispanic origin was

combined with race, and those of Hispanic origin were labeled “Hispanic.” Insurance

status was stratified into private, uninsured/Medicaid, and Medicare/other government.

Facility type was divided into Academic/Research Cancer Programs (ARPs) and non-

ARPs (87). Facility volume was categorized as high or low, with high-volume centers

treating ≥22 patients (≥90th percentile) over the study period. All patients underwent

surgical resection of the primary site, and some received adjuvant radiation or

chemoradiation. Surgical margins were characterized as being negative

(microscopically) or positive (micro- or macroscopically, or unspecified). Patients who

underwent regional lymph node surgery were coded as having received a neck dissection

(87). High-risk pathologic features such as tumor depth, perineural invasion, and

extracapsular extension are not available for the vast majority of patients in the database;

therefore these variables could not be analyzed. Data regarding Charlson comorbidity

index was only available for patients diagnosed years 2003-2006, and was therefore not

examined (2).

Statistical analysis§†

§†Adapted with permissions from:

Schwam ZG, Judson BL. Improved prognosis for patients with oral cavity squamous cell carcinoma: analysis of the

National Cancer Database 1998-2006. Oral Oncol 2016;52:45-51.

Schwam ZG, Sosa JA, Roman SA, Judson BL. Complications and mortality following surgery for oral cavity cancer:

analysis of 408 cases. Laryngoscope 2015;125:1869-73.

18

Statistical analysis was performed with SPSS version 22.0.0 (Chicago, IL).

Standard descriptive statistics were used to summarize patient demographic and disease-

related data. Cases with missing data for variables of interest (<5%) were removed from

the NSQIP cohort. Extreme outliers were removed in calculating summary statistics if

values fell three times the interquartile distance from the 75th percentile value. Chi-square,

Kaplan-Meier, and Cox proportional hazards regression analyses were performed, and

only variables with p≤.10 in univariate analyses were included in multivariate models.

All tests were two-sided, and final threshold for significance was set at p≤.05. Temporal

trends in the first course of treatment and three-year overall survival were depicted using

Prism Graphpad 6.0 (La Jolla, CA), and Pearson product-moment correlation coefficients

were calculated to determine the temporal associations between different treatment

modalities and overall survival. Correlation coefficients range from -1 to +1, which

represent perfect negative and positive linear correlations, respectively. This study was

granted exemption by our institutional review board, as both the NSQIP and NCDB are

de-identified datasets (1,2).

Results

Patient demographics and clinical characteristics in the NSQIP

In querying the NSQIP to identify cases of oral cavity cancer undergoing primary

resection in addition to other procedures, 408 cases were identified. The oral tongue

(40.7%), floor of mouth (21.8%), buccal mucosa (7.1%), retromolar trigone (6.1%), and

hard palate (4.9%) were the most common oral cavity sub-sites (1).

19

The majority of patients were male, between the ages of 45-79 years, and

Caucasian (Table 1). Many had significant comorbid conditions, including diabetes,

tobacco use within the year prior to surgery, heavy alcohol use, prior strokes/transient

ischemic attacks, and hypertension. Few patients received neoadjuvant chemotherapy or

radiotherapy for their disease. Steroids for a chronic condition and poor functional status

prior to surgery were uncommon comorbidities. Weight loss greater than 10% over the

six months prior to surgery occurred in just under 7% of patients, and 3% of patients had

a procedure performed within the 30 days prior to surgery. Neck dissection was

performed in 38.2% of cases (1).

20

Table 1. Patient demographics and clinical characteristics from the NSQIPA

% (N=408)B

Sex

Male 60.0 (245)

Female 40.0 (163)

Age (years)

18-44 7.8 (32)

45-64 43.4 (177)

65-79 34.6 (141)

≥80 14.2 (58)

Race

Caucasian 83.8 (295)

Black 10.2 (36)

Hispanic 2.6 (9)

Asian 3.4 (12)

BMI

<18.5 22.3 (90)

18.5-24.9 38.6 (156)

25.0-29.9 33.7 (136)

≥30.0 5.4 (22)

ASA classification

1 2.5 (10)

2 29.7 (121)

3 63.7 (260)

4 4.2 (17)

A Reproduced with permission from: Schwam ZG, Sosa JA, Roman SA, Judson BL. Complications and mortality

following surgery for oral cavity cancer: analysis of 408 cases. Laryngoscope 2015;125:1869-73. B Numbers may not add to total due to missing data

21

Diabetes 11.8 (48)

Current smoker 32.8 (134)

Heavy alcohol use 9.1 (37)

Pulmonary comorbidity 15.7 (64)

Cardiovascular

comorbidity

11.5 (47)

Hypertension 52.5 (214)

Prior stroke/TIA 8.3 (34)

Neoadjuvant

chemotherapy

1.0 (4)

Neoadjuvant XRT 2.0 (8)

Chronic steroids 3.2 (13)

Coagulopathy 2.7 (11)

Poor functional status 4.2 (17)

Significant weight loss

prior to surgery

6.6 (27)

Operation within prior

30 days

3.0 (12)

Abbreviations: NSQIP=National Surgical Quality Improvement Program, BMI=body mass index,

ASA=American Society of Anesthesiologists, TIA=transient ischemic attack, XRT=radiotherapy

22

30-day postoperative complications and adverse events

The overall thirty-day complication rate was 20.3%, and four patients (1.0%) died within

30 days of surgery (Table 2). The most common adverse events were reoperation as well

as infectious, respiratory, and surgical-site complications. Graft failure, venous

thromboembolism, cardiovascular, and renal complications occurred infrequently.

Unfortunately, detailed information regarding the type of graft (split-thickness skin graft

versus pedicle or free flap) was not available in the database. The majority of

complications occurred between postoperative day (POD) two and ten, although most

venous thromboembolic events and renal insufficiency occurred much later (median 22.0

and 22.5 days, respectively) (1).

23

Table 2. Complications and adverse events within 30 days of surgeryA

% (n)B

Overall 20.3 (83)

Reoperation 9.6 (39)

Infectious 6.6 (27)

Respiratory 5.1 (21)

Surgical-site 4.7 (19)

Flap failure 2.2 (9)

VTE 1.7 (7)

Cardiovascular 1.5 (6)

Renal 1.2 (5)

Death 1.0 (4)

Abbreviations: VTE=venous thromboembolism

A Reproduced with permissions from: Schwam ZG, Sosa JA, Roman SA, Judson BL. Complications and mortality

following surgery for oral cavity cancer: analysis of 408 cases. Laryngoscope 2015;125:1869-73. B Numbers do not add to total due to some patients experiencing multiple complications

24

Preoperative risk factors for 30-day postoperative adverse events

While several preoperative variables were identified as risk factors for

experiencing an adverse event postoperatively, some were linked to a variety of

complications in multivariable analysis (Table 3). Being a smoker within the year prior

to surgery was associated with surgical-site and respiratory complications as well as

reoperation. Undergoing neck dissection was associated with graft failure [odds ratio

[OR] 16.97, 95% confidence interval [CI] 1.78-161.44], surgical-site, infectious, and

respiratory complications, and was associated with an adverse event of any type in

univariate analysis (32.7% versus 12.7%, p<.001). Poor functional status predisposed

patients to infectious and respiratory events, and steroid use for a chronic condition was

the only risk factor for thirty-day mortality (OR 56.44, 95% CI 2.03-1569.28]. There

were no independent risk factors identified for cardiovascular or renal complications (1).

Length-of-stay and post-discharge complications

The median and mean length-of-stay were found to be 3.0 and 4.8 days,

respectively. Patients who experienced an in-hospital adverse event had a significantly

longer inpatient admission (mean 14.8 versus 3.1 days, p<.001). While most adverse

events occurred within the hospital, twenty patients (4.9%) experienced 17.3% of all

complications in the post-discharge time period. A significant proportion of surgical-site

complications (42.1%), sepsis (20.0%), flap failure (22.2%), deep venous thrombosis

(33.3%), and death (25.0%) occurred after admission. While the vast majority (91.3%) of

post-discharge complications occurred by post-discharge day 14, 52% of post-discharge

25

wound dehiscences and 67% of surgical site infections occurred by post-discharge day 7

(1).

26

Table 3. Independent risk factors for postoperative eventsA

Abbreviations: OR=odds ratio, CI=confidence interval, NS=not significant, VTE=venous thromboembolism, Hx=history of, TIA=transient ischemic attack,

ND=neck dissection

A Reproduced with permissions from: Schwam ZG, Sosa JA, Roman SA, Judson BL. Complications and mortality following surgery for oral cavity cancer: analysis of 408 cases.

Laryngoscope 2015;125:1869-73.

Surgical site Infectious VTE Respiratory Reoperation

Risk Factor OR (95% CI) p OR (95% CI) p OR (95% CI) p OR (95% CI) p OR (95% CI) p

Black 3.89 (1.11-13.66) .034 NS NS NS NS

Current smoker 5.13 (1.67-15.76) .004 NS NS 3.59 (1.39-9.29) .008 3.07 (1.53-6.13) .002

Significant weight

loss 4.59 (1.05-20.09) .043 NS NS 3.19 (1.19-8.54) .021

Poor functional

status NS 6.73 (1.51-30.03) .013 NS 5.57 (1.21-25.53) .027 NS

chronic steroid use NS NS 12.96 (1.99-84.47) .007 NS NS

Operation in prior

30 days NS NS 14.31 (2.18-93.83) .006 NS NS

Hx stroke/TIA NS NS NS 3.41 (1.03-11.26) .044 3.68 (1.46-9.29) .006

ND 6.30 (1.84-21.57) .003 3.83 (1.59-9.24) .003 NS 6.17 (2.07-18.35) .001 NS

27

Changes in patient demographics, clinical characteristics, and survival by year of

diagnosis in the NCDB

After applying the several inclusion and exclusion criteria to patient records from

the NCDB, 13,655 patient records were available for analysis. Patients were divided into

two cohorts based on year of diagnosis: those diagnosed 1998-2003 (n=7,734) and those

diagnosed in 2004-2006 (n=5,921). The overall median follow-up time was 43.6 months

(range 0.0-176.2 months), while the median follow-up times for those diagnosed 1998-

2003 and 2004-2006 were 33.9 months (range 0.0-176.2 months) and 55.1 months (0.0-

106.9 months), respectively (2).

Patients diagnosed in the later cohort were younger, more often had private

insurance, and received their care in academic/research facilities and high volume centers

(Table 4). Patients diagnosed in 2004-2006 also had a higher proportion of T≤2 lesions,

N0 nodal disease, and early stage disease (Stages I and II). Negative surgical margins,

neck dissections, and adjuvant chemoradiotherapy were more often administered to

patients diagnosed in 2004-2006 (2).

Observed three-year overall survival improved significantly from 1998 to 2006

(21.7% to 50.9%, p<.001). Patients diagnosed from 2004-2006 were found to have a

median estimated survival of 88.2 months (95% CI 84.2-92.2 months), compared to 52.2

months (95% CI 49.6-54.7 months) in univariate Kaplan-Meier analysis (Figure 1) (2).

28

Table 4. Patient demographics and facility data stratified by year of diagnosisB

1998-2003 2004-2006

% (n=7,734) % (n=5,921) p

Sex

Male 62.6 61.3

.131

Female 37.4 38.7

Age (years)

18-64 50.8 56.4

<.001

≥65 49.2 43.6

Race

White 88.0 88.3

<.001

Black 6.0 4.3

Hispanic 2.1 3.0

Asian 3.4 3.7

Other 0.6 0.7

Insurance

Private 40.5 45.0

<.001

Uninsured/Medicaid 11.9 12.3

Medicare/other

government

47.5 42.6

Facility Type

ARP 42.4 51.7

<.001

Non-ARP 57.6 48.3

Facility Volume

Low 54.9 45.7 <.001

B Reproduced with permissions from Schwam ZG, Judson BL. Improved prognosis for patients with oral cavity

squamous cell carcinoma: analysis of the National Cancer Database 1998-2006. Oral Oncol 2016;52:45-51.

29

High 45.1 54.3

T-classification

1 42.1 45.2

<.001

2 32.2 32.1

3 10.1 7.2

4 15.6 15.5

N-classification

0 78.2 79.6

.006

1 10.7 9.1

2 10.4 10.8

3 0.7 0.5

M-classification

0 99.6 99.6

.846

1 0.4 0.4

Overall Stage

I 39.4 42.4

<.001

II 24.7 25.0

III 13.5 10.7

IV 22.4 21.9

Treatment Regimen

Surgery alone 64.8 66.9

<.001 Surgery + RT 29.7 21.6

Surgery + CRT 5.4 11.5

Surgical Margins

Negative 85.8 88.2

<.001

Positive 14.2 11.8

Neck Dissection

30

No 49.6 47.2

.006

Yes 50.4 52.8

Abbreviations: ARP-Academic/Research Program, RT-Radiation Therapy, CRT-

Chemoradiation Therapy

31

Figure 1. Overall survival of all patients stratified by year of diagnosisB

B Reproduced with permissions from Schwam ZG, Judson BL. Improved prognosis for patients with oral cavity

squamous cell carcinoma: analysis of the National Cancer Database 1998-2006. Oral Oncol 2016;52:45-51.

32

Changes in treatment patterns and survival by year of diagnosis and disease stage

Patients were characterized as having early or late stage (Stages III and IV)

disease, with the proportion of patients being diagnosed with early stage disease

increasing from 64.1% to 67.4% between the two time-based cohorts (p<.001). Patients

presenting with late stage disease were significantly different from early stage patients in

terms of demographics and several treatment variables; patients with later stage disease

were more frequently <65 years of age (57.8% versus 50.9%), uninsured/Medicaid

(18.8% versus 8.6%), non-white (17.3% versus 9.0%), and received their treatment in

high volume centers (59.0% versus 43.9%) (all p<.001) (2).

Treatment patterns for patients with early stage disease remained stable between

1998 and 2006 (Figure 2a); there was a 3.8% increase in the administration of adjuvant

chemoradiotherapy, and a 1.9% decrease in the administration of adjuvant radiotherapy.

Surgery followed by adjuvant radiotherapy had a negative temporal correlation with

three-year overall survival (Pearson coefficient -0.77, p=.016), whereas adjuvant

chemoradiotherapy was positively correlated with three-year overall survival (Pearson

coefficient 0.88, p=.002). There was a significant improvement for patients with early

stage disease in three-year overall survival from 23.5% to 59.7% over the study period

(p<.001) (2).

Early stage disease patients diagnosed in 1998-2003 had higher rates of positive

margins (11.0% versus 8.3%), treatment in a low volume center (61.1% versus 50.0%),

33

and age ≥65 years (52.1% versus 45.4%) (all p<.001). Neck dissections increased from

22.8% to 37.6% over the study period (p<.001) (2).

Patients with late stage disease saw a similar improvement in three-year overall

survival from 17.9% to 33.9% over the study period (Figure 2b). A strong, positive

temporal correlation was found between administration of adjuvant chemoradiotherapy,

which increased from 8.3% to 36.4%, and three-year overall survival (Pearson coefficient

0.92, p<.001). Receipt of adjuvant radiotherapy was observed to fall from 54.0% to

31.4%, and was negatively associated with survival (Pearson coefficient=-0.95, p<.001)

(2).

Late stage disease patients diagnosed from 2004 to 2006 were more often <65

years old (60.2% versus 56.0%, p=.004), more commonly had private insurance (40.9%

versus 37.5%, p=.008), received treatment in a high volume facility (63.1% versus 56.1%,

p<.001), and presented with Stage IV disease (67.0% versus 62.4%, p=.001). No

statistical difference was found in surgical margin status between the two periods

(p=.341) (2).

34

Figure 2a. Temporal trends in treatment and three-year overall survival among patients

with early stage diseaseB

Abbreviations: RT-radiation therapy, CRT-chemoradiation therapy, OS-overall survival

B Reproduced with permission from Schwam ZG, Judson BL. Improved prognosis for patients with oral cavity

squamous cell carcinoma: analysis of the National Cancer Database 1998-2006. Oral Oncol 2016;52:45-51.

1998 2000 2002 2004 20060

10

20

30

40

50

60

70

80

90

100

Year of Diagnosis

Perc

ent

Stages I + II Disease

Surgery

Surgery + RT

Surgery + CRT

3-yr OS

35

Figure 2b. Temporal trends in treatment and three-year overall survival among patients

with late stage diseaseB

Abbreviations: RT-radiation therapy, CRT-chemoradiation therapy, OS-overall survival

B Reproduced with permission from Schwam ZG, Judson BL. Improved prognosis for patients with oral cavity

squamous cell carcinoma: analysis of the National Cancer Database 1998-2006. Oral Oncol 2016;52:45-51.

1998 2000 2002 2004 20060

10

20

30

40

50

60

Year of Diagnosis

Perc

ent

Stages III + IV Disease

Surgery

Surgery + RT

Surgery + CRT

3-yr OS

36

Long-term prognostic factors in multivariable analysis

Several independent risk factors for mortality were identified in multivariable

analysis. Older age, insurance status, year of diagnosis, later stage at diagnosis,

administration of adjuvant chemoradiotherapy, and the presence of positive margins were

all associated with increased mortality in early and late stage disease patients, but also

when these two disease-based cohorts were combined (Table 5). Receiving treatment in

a high volume facility and undergoing neck dissection were associated with better overall

survival in early stage disease patients and when all patients were combined, but not in

late stage disease patients. Being treated in an academic/research program was linked to

improved prognosis when all patients were combined, but not for a particular staging

group. Compared to surgery alone, receiving adjuvant radiotherapy was associated with

greater risk of death in early stage disease patients and when all patients were combined.

The year of diagnosis was associated with improved overall survival (Hazard ratio [HR]

0.76, 95% CI 0.72-0.80) in multivariable models, and those diagnosed 2004-2006 had a

significantly higher three-year overall survival rate as compared to those diagnosed in the

earlier cohort (48.4% versus 29.2%, p<.001) (2).

37

Table 5. Independent risk factors for mortality stratified by disease-stateB Early Stage (I, II) Late Stage (III, IV) All Stages (I-IV)

HR (95% CI) p HR (95% CI) p HR (95% CI) p

Sex

Male (Ref) 1.00 1.00 1.00

Female 1.00 (0.93-1.07) .943 0.92 (0.85-0.99) .020 0.95 (0.91-1.01) .062

Age (years)

18-64 (Ref) 1.00 1.00 1.00

≥65 1.80 (1.62-1.99) <.001 1.41 (1.26-1.58) <.001 1.59 (1.48-1.72) <.001

Race

White (Ref) 1.00 1.00 1.00

Black 1.43 (1.21-1.69) <.001 0.99 (0.87-1.12) .866 1.10 (0.99-1.21) .078

Hispanic 1.00 (0.79-1.69) .998 0.85 (0.66-1.08) .173 0.91 (0.77-1.08) .271

Asian 0.71 (0.57-0.89) .003 0.85 (0.70-1.02) .075 0.78 (0.68-0.90) .001

Other 0.96 (0.59-1.54) .859 0.87 (0.57-1.32) .503 0.88 (0.64-1.21) .444

Insurance

Private (Ref) 1.00 1.00 1.00

Uninsured/Medicaid 1.54 (1.35-1.76) <.001 1.34 (1.32-1.49) <.001 1.52 (1.40-1.64) <.001

Medicare/other

government

1.58 (1.43-1.76) <.001 1.29 (1.15-1.44) <.001 1.45 (1.35-1.57) <.001

Facility Type

ARP (Ref) 1.00 1.00 1.00

Non-ARP 1.05 (0.97-1.15) .121 1.08 (0.98-1.18) .123 1.07 (1.01-1.14) .030

Facility Volume

Low (Ref) 1.00 1.00 1.00

B Reproduced with permission from Schwam ZG, Judson BL. Improved prognosis for patients with oral cavity

squamous cell carcinoma: analysis of the National Cancer Database 1998-2006. Oral Oncol 2016;52:45-51.

38

High 0.84 (0.79-0.90) <.001 0.97 (0.88-1.06) .470 0.91 (0.85-0.97) .002

Year of Diagnosis

1998-2003 (Ref) 1.00 1.00 1.00

2004-2006 0.67 (0.63-0.72) <.001 0.87 (0.81-0.94) <.001 0.76 (0.72-0.80) <.001

Stage

I (Ref) 1.00 -- -- 1.00

II 1.49 (1.39-1.60) <.001 -- -- 1.49 (1.40-1.60) <.001

III (Ref) -- -- 1.00 1.96 (1.80-2.14) <.001

IV -- -- 1.22 (1.13-1.32) <.001 2.44 (2.26-2.64) <.001

Treatment Regimen

Surgery (Ref) 1.00 1.00 1.00

Surgery + RT 1.39 (1.38-1.51) <.001 1.00 (0.92-1.08) .999 1.18 (1.11-1.25) <.001

Surgery + CRT 1.80 (1.48-2.18) <.001 1.13 (1.02-1.26) .021 1.41 (1.29-1.54) <.001

Neck Dissection

No (Ref) 1.00 1.00 1.00

Yes 0.80 (0.75-0.87) <.001 1.00 (0.91-1.11) .952 0.90 (0.84-0.95) <.001

Surgical Margins

Negative (Ref) 1.00 1.00 1.00

Positive 1.39 (1.26-1.53) <.001 1.55 (1.43-1.69) <.001 1.50 (1.41-1.60) <.001

Abbreviations: HR-Hazard Ratio, CI-Confidence interval, Ref-Referent category, ARP-

Academic/Research Program, RT-Radiation Therapy, CRT-Chemoradiation Therapy

39

Factors associated with receipt of adjuvant therapies

When all patients were combined, 65.7% received surgery alone, 29.1% received

adjuvant radiotherapy, and 8.1% received adjuvant chemoradiotherapy. In univariate

analysis, later year of diagnosis, younger age, black race, private insurance, getting

treated in an academic program or high volume facility, T4 lesions, higher overall stage,

positive margins, and undergoing neck dissection were all associated with the receipt of

adjuvant chemoradiotherapy (Table 6) (2).

Nearly one in six patients receiving adjuvant chemoradiotherapy did not undergo

neck dissection, and 76.1% had negative surgical margins. Data regarding dose of

radiotherapy were scant, being present for only 21.1% of our cohort. In those patients for

whom data were available, the median dose was 56.0 Gray (Gy), and 18.2% received

doses between 60.0 and 66.0 Gy. The median dose increased over the study period from

52.0 Gy (1998-2003) to 59.4 Gy (2004-2006). No data were available regarding type,

dose, or adequacy of chemotherapeutic agents (2).

40

Table 6. Unadjusted risk factors for administration of adjuvant therapiesB

Surgery alone Surgery + RT Surgery + CRT p

% (n=8,979) % (n=3,974) % (n=1,102)

Year of Diagnosis

1998-2003 55.9 64.3 38.2

<.001

2004-2006 44.1 35.7 61.8

Sex

Male 60.9 63.5 66.4

.001

Female 39.1 36.5 33.6

Age (years)

18-64 48.6 57.8 76.1

<.001

≥65 51.4 42.2 23.9

Race

White 90.6 84.4 79.9

<.001

Black 3.5 8.1 10.0

Hispanic 2.1 2.9 3.9

Asian 3.2 3.9 5.2

Other 0.6 0.6 1.0

Insurance

Private 40.9 42.8 54.3

<.001

Uninsured/Medicaid 10.0 15.2 19.7

Medicare/other

government

49.1 42.0 26.0

Facility Type

ARP 44.1 48.9 57.7 <.001

B Reproduced with permission from Schwam ZG, Judson BL. Improved prognosis for patients with oral cavity

squamous cell carcinoma: analysis of the National Cancer Database 1998-2006. Oral Oncol 2016;52:45-51.

41

Non-ARP 55.9 51.1 42.3

Facility Volume

Low 51.5 51.9 42.6

<.001

High 48.5 48.1 57.4

T-classification

1 56.7 19.7 12.6

<.001

2 29.2 39.3 32.6

3 5.3 14.6 18.7

4 8.8 26.4 35.1

N-classification

0 90.1 62.9 38.5

<.001

1 5.5 17.9 21.2

2 4.1 18.6 37.3

3 0.2 0.7 3.0

M-classification

0 99.7 99.6 99.1

.003

1 0.3 0.4 0.9

Overall Stage

I 55.2 15.2 5.8

<.001

II 25.4 26.5 14.4

III 7.9 20.6 20.9

IV 11.4 37.7 58.9

Surgical Margins

Negative 92.2 76.8 76.1

<.001

Positive 7.8 23.2 23.9

Neck Dissection

No 60.0 30.0 15.9 <.001

42

Yes 40.0 70.0 84.1

Abbreviations: RT-radiation therapy, CRT-chemoradiation therapy, ARP-

Academic/Research Program

Discussion

There is a scarcity of multi-institutional level data characterizing the postoperative

complications following oral cavity cancer surgery, the risk factors for them, and the

time-course over which they occur (1). While changes in prognosis have been reported

in oral cancer in international cohorts (81), there is a similar lack of multi-institutional

data from the United States (2).

In measuring thirty-day morbidity and mortality following primary resection for

oral cavity cancer, the overall complication rate was 20.3% and mortality occurred in

1.0% of patients. In their comparison of clinical registry and administrative data from

355 oral cavity cancer patients treated at Memorial Sloan-Kettering Cancer Center, Awad

et al. found that all major complications, one-third of total complications, and 27% of

patients were found in the NSQIP, in comparison to their own administrative records (90).

They found a significantly higher overall adverse rate of 62.0%, but were also able to

track many procedure-specific complications including orocutaneous fistula, trismus, and

chyle leak that are not found in the NSQIP. Awad et al. tracked only complications

occurring during admission, and stopped tracking them after 45 days (1).

43

Awad’s group found that 34% of patients undergoing neck dissection experienced

a complication (90), while we measured a very similar value of 33%. The rate of

complications after neck dissection ranges in the literature between 7-16% for major

complications and up to 38% in total (110,111,112,113,114), although pre-neck

dissection therapy such as radiation, chemotherapy, or prior surgery must be taken into

consideration in addition to extent of neck dissection. In one prospective trial of elective

modified radical (MRND) versus supraomohyoid (SOH) neck dissection for T2-T4 oral

cavity malignancies, the MRND group had a significantly higher complication rate (41%

versus 25%, p=.043) and a longer median length of stay (9 days versus 7 days) when

compared to the SOH group (115). In our study, we found that undergoing neck

dissection was a risk factor for several postoperative complications, including surgical-

site, infectious, and respiratory complications. While the neck dissection itself may

increase the likelihood of complications by prolonging operative times, extending

incisions, and putting critical neurovascular structures at risk, we believe it to be a proxy

variable for more advanced disease (37) in the NSQIP, which does not track oncologic

variables such as TNM classifications or overall stage of disease (1).

We found that the rate of neck dissections increased significantly between 1998

and 2006, particularly for patients with early stage disease. It was also noted that

undergoing neck dissection was associated with better overall survival in early stage

disease patients and when all patients were combined (2). A recent randomized control

trial from Tata Memorial Centre evaluated the role of elective versus therapeutic neck

dissection in 596 patients with T≤2, clinically N0 lateralized oral cancers (47). The

44

investigators found several benefits to elective node dissection, including significantly

fewer recurrences, a three-year overall survival rate of 80.0% compared to 67.5%, a

hazard ratio for mortality of 0.64 (95% CI 0.45-0.92), and a significantly higher rate of

three-year disease-free survival (69.5% versus 45.9%). Adverse event rates were

relatively low, with 6.6% of the elective group and 3.6% of the therapeutic group having

complications.

In addition to neck dissection as a significant risk factor for multiple

complications, we found that cigarette smoking within the year prior to surgery and

weight loss ≥10% within the six months prior to surgery were associated with surgical-

site complications. Perioperative smoking cessation is frequently recommended due to its

known association with impaired wound healing (116,117,118) and surgical site

infections (1,119,120). Weight loss resulting in physiologic impairments has been

associated with postoperative septic complications (1,121). In our study, we believe that

significant weight loss most likely reflects cancer-induced inflammation as well as

difficulties with eating as a result of the lesion’s location in the oral cavity (1).

Furthermore, it has been shown that systemic inflammation and weight loss affect

albumin concentration (122,123), which was associated with wound complications in a

NSQIP study of over two thousand patients undergoing laryngectomy (1,124).

Minimizing readmission after surgery has become a quality of care initiative

(125), especially in light of its ability to influence Medicare reimbursements (126).

While the NSQIP does not track readmission rates, Luryi et al. found a 3.5% readmission

45

rate following surgery for oral cancers (87) using the National Cancer Database. In the

same study, neck dissection, male gender, and T3 tumor classification were associated

with unplanned readmission. In our cohort, we measured thirty-day reoperation

occurring in 9.6% of patients, with risk factors including cigarette smoking, recent weight

loss, and history of stroke/TIA. As these very same risk factors were also noted to be

associated with surgical-site and respiratory complications, reoperation may have been to

create or revise a tracheostomy stoma. Cigarette smoking in the two weeks prior to

surgery has also been linked to major flap complications requiring return to the operating

room (127), and may explain some of the reoperations in our study (1).

Post-discharge complications are a huge source of hospital readmissions, with

nearly one in seven surgical patients experiencing a preventable adverse event leading to

unintended readmission (105,128). In our study, we found that 4.9% of patients

experienced 17.3% of all complications after leaving the hospital, and that over 90% of

them occurred within 14 days of discharge. Many superficial wound infections and much

of the dehiscence occur within 7 days of discharge, however. This compares favorably to

the NSQIP analysis of 21 groups of general surgery procedures by Kazaure and

colleagues, in which 41.5% of all complications occurred in the post-discharge timeframe

(105). In using the NSQIP to examine over 48,000 patients undergoing a diverse range

of otolaryngologic procedures, Chen et al. had similar findings, showing that about three-

quarters of post-discharge complications occurred by post-discharge day 14, with many

of them concerning the surgical-site or systemic infections (1,88).

46

In order to maximize short-term well-being and to minimize healthcare spending

on unintended readmissions, the literature shows that early clinical follow-up is essential

(1). It has been reported that half of patients readmitted for post-discharge complications

had not seen a physician within the month after discharge (128). Using the Surveillance,

Epidemiology, and End Results (SEER) database, Tuggle et al. showed that seeing a

healthcare provider in the 30 days after surgery could reduce re-hospitalization rates by

55% (absolute decrease of 6.3%) in elderly patients undergoing thyroidectomy (129).

Similarly, Robles was able to show that hiring a nurse practitioner in a surgical practice

not only increased telephone communication and therapy services, but also reduced

unnecessary emergency room visits by 48% (absolute reduction of 12%) (130). As many

of the post-discharge complications occurred within the one to two weeks after leaving

the hospital, early intervention would likely play an important role in minimizing

morbidity and unnecessary healthcare spending (1).

Our data from the National Cancer Database demonstrate a large increase in

three-year overall survival for patients with oral cavity squamous cell carcinomas using

multi-institutional data from hundreds of participating centers throughout the United

States. In a 20-year retrospective study of 2,738 oral cavity cancer patients from seven

leading international cancer centers, Amit et al. found an 11% improvement in five-year

overall survival and a 12% increase in disease-specific survival when all patients were

examined together (81). Surgery alone was performed in 31% of their cohort, adjuvant

radiotherapy was administered in 51%, and adjuvant chemoradiotherapy in 18%.

Additionally, 60% of patients underwent elective neck dissection, and 40% received

47

therapeutic neck dissection. When splitting the study population into two time-of-

diagnosis based cohorts (1990-2000 and 2001-2011), the authors found that the later

cohort presented at an earlier age, with more advanced disease and extracapsular

extension, and more often had negative margins. A limited, selective neck dissection was

also more frequently performed in the later group. The decade of diagnosis was an

important predictor of survival, with patients diagnosed 2001-2011 having better overall,

disease-specific, and disease-free survival in multivariate analyses. In a study by Pulte et

al. using the SEER database to examine trends in prognosis of head and neck cancers, the

investigators similarly found a 14.4% increase in overall survival for patients with lesions

of the oral tongue between 1992-1996 and 2002-2006 (131). We found a much larger

improvement in overall survival than found in the Amit and Pulte studies, which may be

due to either changes in patterns of follow-up by the NCDB or heterogeneity of the

participating sites; the NCDB includes hospital registry data from over 1500 participating

sites (97), while Amit and colleagues tracked patients only from seven leading cancer

hospitals throughout the world. We found period of diagnosis to be associated with

improved overall survival, and found that those presenting later more often were younger

and had early stage disease; this may be explained by greater access to healthcare, more

effective disease screening, or increased awareness (2).

Receiving treatment in a high-volume center took place more often in 2004-2006,

and was associated with improved overall survival in patients with early stage disease

and when all patients were analyzed together. This was closely mirrored by an increase

in patients receiving care in academic/research programs. These changes in treatment

48

setting may be from some degree of regionalization of cancer care, shifts in referral

patterns, or more physicians going to work for larger hospital systems (2). Facility

volume has been shown to be a positive factor in head and neck cancer prognostication

using multiple databases (133,134) and government registries (135,136). Surgical

resection of the primary tumor is often the first therapeutic step in the treatment of oral

cavity malignancies (37), and higher-volume centers may be more facile with larger

ablative procedures requiring free tissue transfer as part of the reconstruction. Higher

volume centers may also have a more formal way of approaching oncologic treatment

decision making in the way of tumor boards, which have been shown to intensify

proposed head and neck cancer treatment and up-stage disease (2,136). We found that

nearly 80% of high volume facilities were academic/research programs, which have been

largely superior to non-academic facilities in terms of process measures and risk-adjusted

mortality for complex patients in the literature (137). We labeled a facility “high volume”

if it was in the ≥90th percentile, and did not determine the minimum percentile for which

there was a survival benefit (2).

In addition to treatment in a high volume facility, this study found positive

surgical margins to be a poor prognostic factor for patients with oral cavity squamous cell

carcinoma. Positive margins are more commonly found in the oral cavity when

compared to other head and neck sub-sites (138), are an important determinant in the

decision to administer adjuvant treatment (37), and put patients at higher risk of disease

recurrence and compromised disease-specific survival (2,139). We found a significant

decrease in the rate of patients with positive margins and a steady increase in patients

49

receiving adjuvant therapy; this is likely due to high-risk pathologic features not

accounted for in the database, such as extracapsular spread and perineural invasion. In

using the NCDB to examine over 20,000 early stage oral cavity cancer patients, Luryi et

al. observed positive margins in 7.5% of the study population, with a range of 0.0-43.8%

(140). The authors found that more advanced disease and receiving treatment in a lower

volume or non-academic facility to be risk factors for positive surgical margins. While

we did not examine risk factors for positive margins, we did see a corresponding decrease

in both presentation with late stage disease and treatment in low volume/non-academic

facilities to accompany the decrease in rate of positive margins (2).

Receipt of adjuvant chemoradiotherapy was associated with worse overall

survival in all patients, regardless of disease stage. This is in contrast to its strong,

positive temporal correlation with three-year overall survival in late stage patients.

Administration of adjuvant chemoradiotherapy became increasingly popular after the

publication of two noteworthy trials, RTOG 9501 (66) and EORTC 22931 (67), which

compared the efficacy of adjuvant chemoradiotherapy to adjuvant radiotherapy in high-

risk and late stage head and neck cancer, respectively. Both studies reported superior

locoregional control and a more serious toxicity profile in the chemoradiotherapy group.

The two trials differed slightly with respect to survival analyses; RTOG 9501 reported

that chemoradiotherapy was associated with improved disease-free survival, but not

overall survival, while EORTC 22931 demonstrated better progression-free and overall

survival in the chemoradiotherapy group. When the authors pooled their data in a follow-

up analysis, they concluded that adjuvant chemoradiotherapy was best suited for those

50

with extracapsular extension or positive margins (2,65). In our cohort, nearly one in six

patients receiving adjuvant chemoradiotherapy did not undergo neck dissection, and over

three-quarters receiving such therapy had negative margins. Additionally, many patients

with positive margins did not receive adjuvant therapy. While we were unable to account

for certain high-risk pathologic features that are currently indications for adjuvant

treatment, the appropriateness of certain treatment regimens must be called into question

(2). Unfortunately, the reason for non-administration of therapy was lacking for the

majority of patients, and could not be analyzed in a meaningful way. It is possible that

patients had significant comorbidities precluding them from receiving treatment, or that

they refused recommended therapy. The majority of patients were also diagnosed and

treated prior to the publication of those two trials. In analyzing the adequacy of

radiotherapy, we found that the median regional dose to be 56.0 Gy, with only 18.2% of

patients for whom there was data receiving doses in line with the trial protocols.

Adequacy of chemotherapy or systemic therapies was not possible to measure in the

database, and toxicity from adjuvant therapy as a cause for mortality cannot be excluded

(2).

Several additional limitations must be taken into account when analyzing the

results of our data concerning both short- and long-term prognosis of oral cavity

malignancies. In terms of thirty-day morbidity and mortality, many complications of

interest may not be tracked in the NSQIP, potentially underestimating the true

complication rate. Patients in our cohort also underwent a variety of concurrent

procedures, from free flaps to parotidectomies, with each procedure having a different

51

adverse event profile. Furthermore, the NSQIP does not keep track of oncologic

variables, hospital or surgeon volume, prior surgery, or rationale/timing of reoperation (1).

Several other variables were not adjusted for, and may impact our results: including

trainee involvement, length of procedure, surgeon skill, postoperative nursing care, and

patient compliance with prescribed regimens. While we were able to give a timecourse

over which complications occurred, there is wide variation in admission and discharge

practices amongst facilities. An important variable not recorded in the NSQIP is time

from hospital discharge to outpatient follow-up, as this may be an important avenue for

intervention. Additionally, some complications in the NSQIP may not be as specific as

desired; for example, return to the operating room is listed, but without cause. Returning

because of postoperative hemorrhage and returning to obtain negative margins are very

different events. Additionally, we selected a cohort in which data for the variables of

interest were present, introducing potential bias, unless one assumes that data is missed at

random. While the NSQIP collects data from its hundreds of participating sites, entry

requirements including a large subscription fee and the cost of employing a SCNR; this

may be prohibitive for smaller facilities or may skew the dataset to representing

institutions with a strong desire to study and improve outcomes.

In analyzing changes in long-term prognosis, we were unable to analyze Charlson

comorbidity scores, relative timing of adjuvant treatment, and a host of high-risk

pathologic features that are associated with worse prognosis in oral cancers. The NCDB

also limits analysis to overall survival, as cause of death is not available. Individual

patient income and level of education are not available, which may impact treatment

52

decisions. Undergoing neck dissection was not one of our inclusion criteria, which may

have led to under-staging for certain patients. We limited our analysis to patients

diagnosed between 1998 and 2006 due to data availability; if extended in either direction,

a more pronounced change in prognosis may have been observed (2). Additionally, the

NCDB has additional limitations that should further inform the results of our study. As it

currently stands, only CoC-approved cancer centers are allowed to report their results to

the NCDB (141,142); criteria for CoC-approval include having a wide range of oncologic

services and specialists, and must participate in several quality-improvement activities

(141). CoC-approved hospitals have been found to be more likely associated with

residency programs and medical schools, more frequently accredited by the Joint

Commission, less frequently in rural areas, and tend to have higher volume surgical

facilities and more comprehensive oncologic services (141). As the vast majority of

facilities are not CoC-accredited (142), the NCDB may represent an unrepresentative

cross-section of oral cancer care nationwide.

Conclusions

In examining a multicenter cohort of patients with oral cavity cancer, we

determined that a significant number of patients undergoing primary resection experience

a postoperative adverse event, and determined that a small yet significant minority

experience complications after admission. We identified risk factors that may serve as

potential red flags in the patient history for postoperative complications. Furthermore, we

identified the two weeks after discharge as a high-risk time for vulnerable patients, and

believe that early clinical follow-up may mitigate the morbidity associated with post-

53

discharge complications and prevent costly readmission in high-risk patients (1).

Monitoring our advancements in treatment of oncologic disease is an important task to

undertake, and can be done through examining changes in prognosis for a given disease.

This was the first study showing that the long-term prognosis for patients with oral cavity

squamous cell carcinoma has improved significantly for both early and late stage disease

patients using a nationwide cohort. Similar changes in prognosis have only been

demonstrated in patients with human papillomavirus-positive oropharyngeal squamous

cell carcinoma. This study was the first to show that receipt of adjuvant

chemoradiotherapy has increased substantially in the treatment of oral cavity squamous

cell carcinoma, yet we also found it to be associated with compromised overall survival.

We think this represents a selection bias where patients with high-risk pathologic features

not accounted for in the NCDB were more likely to receive such therapy. We believe

that further work is needed in the way of clinical trials to determine the efficacy of

adjuvant therapies specifically in patients with oral cavity squamous cell carcinoma (2).

54

References

1. Schwam, Z.G., Sosa, J.A., Roman, S.A., Judson, B.L. 2015. Complications and

mortality following surgery for oral cavity cancer. Laryngoscope. 125:1869-73.

2. Schwam, Z.G., Judson, B.L. 2016. Improved prognosis for patients with oral cavity

squamous cell carcinoma: Analysis of the National Cancer Database 1998-2006. Oral

Oncol. 52:45-51.

3. Wein, R.O., Weber, R.S. 2014. Malignant neoplasms of the oral cavity. In Cummings

Otolaryngology: P.W. Flint, B.H. Haughey, V. Lund, J.K. Niparko, T. Robbins, J.R.

Thomas, M.M. Lesperance, editors. Philadelphia: Saunders. 1359-1387.

4. Shah, J.P. 1990. Patterns of cervical lymph node metastases from squamous cell

carcinomas of the upper aerodigestive tract. Am. J. Surg. 160:405-409.

5. Siegel, R.L., Miller, K.D., Jemal, A. 2015. Cancer statistics, 2015. CA: A cancer

journal for clinicians. 65:5-29.

6. World Health Organization Global Agency for Research on Cancer. 2012. World

estimated cancer incidence, all ages: both sexes. Available from URL:

http://globocan.iarc.fr/old/summary_table_pop-

html.asp?selection=224900&title=World&sex=0&type=0&window=1&sort=0&submit=

%C2%A0Execute%C2%A0 [accessed July 4, 2015].

55

7. Warnakulsariya, S. 2009. Global epidemiology of oral and oropharyngeal cancer. Oral

Oncol. 45:309-16.

8. Llewellyn, C.D., Johnson, N.W., Warnakulasuriya, S. 2001. Risk factors for squamous

cell carcinoma of the oral cavity in young people – a comprehensive literature review.

Oral Oncol. 37:401–418.

9. Surveillance epidemiology and end results (SEER). SEER Cancer statistics review

1975–2004. National Cancer Institute. Available from URL:

http://seer.cancer.gov/statfacts/html/oralcav [accessed November 1, 2015].

10. Shiboski, C.H., Schmidt, B.L., Jordan, R.C. 2005. Tongue and tonsillar carcinoma:

increasing trends in the US population ages 20–44 years. Cancer. 103:1843–1849.

11. Chaturvedi, A.K., Engels, E.A., Anderson, W.F., Gillison, M.L. 2008. Incidence

trends for human papillomavirus–related and –unrelated oral squamous cell carcinomas

in the United States. J. Clin. Oncol. 26:612-619.

12. Winn, D.M., Blot, W.J., Shy, C.M., Pickle, L.W., Toledo, A., et al. 1981. Snuff

dipping and oral cancer among women in the southern United States. N. Engl. J. Med.

304:745–9.

56

13. Decker, J., Goldstein, J.C. 1982. Risk factors in head and neck cancer. N. Engl. J.

Med. 306:1151-1155.

14. Reddy, C.R. 1974. Carcinoma of hard palate in India in relation to reverse smoking of

chuttas. J. Natl. Cancer Inst. 53:615-9.

15. Franceschi, S., Talamini, R., Barra, S., Baron A.E., Negri, E. 1990. Smoking and

drinking in relation to cancers of the oral cavity, pharynx, larynx, and esophagus in

northern Italy. Cancer. Res. 15:6502-7.

16. Wynder, E.L., Hoffmann, D. 1976. Tobacco and tobacco smoke. Semin. Oncol. 3:5–

15.

17. Hecht, S.S. 2003. Tobacco carcinogens, their biomarkers and tobacco-induced cancer.

Nat. Rev. Cancer. 3:733-44.

18. Mansour, O.I., Snyderman, C.H., D’Amico, F. 2003. Association between tobacco

use and metastatic neck disease. Laryngoscope. 113:161-166.

19. Seitz, H.K., Stickel, F. 2007. Molecular mechanisms of alcohol-mediated

carcinogenesis. Nat. Rev. Cancer. 7:599-612.

57

20. Mashberg, A., Garfinkel, L., Harris, S. 1981. Alcohol as a primary risk factor in oral

squamous cell carcinoma. CA Cancer J. Clin. 31:146-55.

21. Rothman, K., Keller, A. 1972. The effect of joint exposure to alcohol and tobacco on

risk of cancer of the mouth and pharynx. J. Chronic Dis. 25:711–6.

22. Homann, N., Jousimies-Somer, H., Jokelainen, K., Heine, R., Salaspuro, M. 1997.

High acetaldehyde levels in saliva after ethanol consumption: methodological aspects and

pathogenetic implications. Carcinogenesis. 18:1739–1743.

23. Seitz, H.K., Stickel, F. 2006. Risk factors and mechanisms of hepatocarcinogenesis

with special emphasis on alcohol and oxidative stress. Biol. Chem. 387:349–360.

24. Garro, A.J., Espina, N., Farinati, F., Lieber, C.S. 1986. The effect of chronic ethanol

consumption on carcinogen metabolism and on O6-methylguanine transferase-mediated

repair of alkylated DNA. Alcohol Clin. Exp. 10:73–77.

25. Wang, M., McIntee, E.J., Cheng, G., Shi, Y., Villalta, P.W., et al. 2000. Identification

of DNA adducts of acetaldehyde. Chem Res Toxicol. 13:1149–1157.

26. Bailey, S.M., Cunningham, C.C. 2002. Contribution of mitochondria to oxidative

stress associated with alcohol liver disease. Free. Radic. Biol. Med. 32, 11–16.

58

27. De Visscher, J.G., Van der Waal, I. 1998. Etiology of cancer of the lip: A review. Int.

J. Oral Maxillofac. Surg. 27:199-203.

28. Van Leeuwen, M.T., Grulich, A.E., McDonald, S.P., McCredie, M.R., Amin, J, et al.

2009. Immunosuppression and other risk factors for lip cancer after kidney

transplantation. Cancer Epidemiol. Biomarkers Prev. 18:561-9.

29. Wurman, L.H., Adams, G.L., Meyerhoff, W.L. 1975. Carcinoma of the lip. Am. J.

Surg. 130:470-474.

30. Czerninski, R., Zini, A., Sgan-Cohen, H.D. 2010. Lip cancer: incidence, trends,

histology and survival: 1970-2006. Br. J. Dematol. 162:1103-9.

31. Velly, A.M., Franco, E.L., Schlecht, N., Pintos, J., Kowalski, L.P., et al. 1998.

Relationship between dental factors and risk of upper aerodigestive tract cancer. Oral

Oncol. 34:284-291.

32. Marshall, J.R., Graham, S., Haughey, B.P., Shedd, D., O’Shea, R., et al. 1992.

Smoking, alcohol, dentition and diet in the epidemiology of oral cancer. Oral Oncol.

Euro. J. Cancer. 28B:9-15.

59

33. Funk, G.F., Karnell, L.H., Robinson, R.A., Zhen, W.K., Trask, D.K., et al. 2002.

Presentation, treatment, and outcome of oral cavity cancer: a National Cancer Data Base

report. Head Neck. 24:165-180.

34. Spector, M.E., Wilson, K.F., Light, E.L., McHugh, J.B., Bradford, C.R. 2011.

Clinical and pathologic predictors of recurrence and survival in spindle cell carcinoma.

Otolaryngol. Head Neck Surg. 145:242-247.

35. Raslan, W.F., Barnes, L., Krause, J.R. 1994. Basaloid squamous cell carcinoma of the

head and neck: a clinicopathologic and flow cytometric study of 10 new cases with

review of the English literature. Am. J. Otolaryngol. 15:204-211.

36. de Sampaio Goes, F.C., Oliviera, D.T., Dorta, R.G., Nishimoto, I.N., Landman, G., et

al. 2004. Prognoses of oral basaloid squamous cell carcinoma and squamous cell

carcinoma: a comparison. Arch. Otolaryngol. Head Neck Surg. 130:83-86.

37. Pfister, D.G., Spencer, S., Brizel, D.M., Burtness, B., Busse, P.M., et al. NCCN

Clinical Practice Guidelines in Oncology (NCCN Guidelines®) Head and Neck Cancers

Version 1.2015. ©2015 National Comprehensive Cancer Network, Inc. Available from

URL: www.NCCN.org [accessed February 3, 2015].

38. Fakih, A.R., Rao, R.S., Borges, A.M., Patel, A.R. 1989. Elective versus therapeutic

neck dissection in early carcinoma of the oral tongue. Am. J. Surg. 158:309-313.

60

39. Ferlito, A., Rinaldo, A., Robbins, K.T., Leemans, C.R., Shah, J.P., et al. 2003.

Changing concepts in the surgical management of the cervical node metastasis. Oral

Oncol. 39:429-435.

40. Teichgraeber, J.F., Clairmont. A.A. 1984. The incidence of occult metastases for

cancer of the oral tongue and floor of the mouth: treatment rationale. Head Neck Surg.

7:15-21.

41. Ross, G.L., Soutar, D.S., MacDonald, D.G., Shoaib, T., Camilleri, I.G., et al. 2004.

Improved staging of cervical metastases in clinically node-negative patients with head

and neck squamous cell carcinoma. Ann. Surg. Oncol. 11:213-218.

42. Keski-Santti, H., Atula, T., Tornwall, J., Koivunen, P., Makitie, A. 2006. Elective

neck treatment versus observation in patients with T1/T2 N0 squamous cell carcinoma of

oral tongue. Oral Oncol. 42:96-101.

43. Pentenero, M., Gandolfo, S., Carrozzo, M. 2005. Importance of tumor thickness and

depth of invasion in nodal involvement and prognosis of oral squamous cell carcinoma: a

review of the literature. Head Neck. 27:1080-1091.

44. Moore, C., Kuhns, J.G., Greenberg, R.A. 1986. Thickness as prognostic aid in upper

aerodigestive tract cancer. Arch. Surg. 121:1410-1414.

61

45. DiTroia, J.F. Nodal metastases and prognosis in carcinoma of the oral cavity. 1972.

Otolaryngol. Clin. North Am. 5:333-342.

46. Huang, S.H., Hwang, D., Lockwood, G., Goldstein, D.P., O'Sullivan, B. 2009.

Predictive value of tumor thickness for cervical lymph‐node involvement in squamous

cell carcinoma of the oral cavity. Cancer. 11:1489-1497.

47. D’Cruz, A.K., Vaish, R., Kapre, N., Dandekar, M., Gupta, S., et al. 2015. Elective

versus therapeutic neck dissection in node-negative oral cancer. N. Engl. J. Med.

373:521-529.

48. Chandu, A., Smith, A.C., Rogers, S.N. 2006. Health-related quality of life in oral

cancer: a review. J. Oral. Maxillofac. Surg. 64:495-502.

49. Rogers, S.N., Lowe, D., Brown, J.S., Vaughan, E.D. 1999. The University of

Washington head and neck cancer measure as a predictor of outcome following primary

surgery for oral cancer. Head Neck. 21:394-401.

50. De Graeff, A., de Leeuw, J.R., Ros, W.J., Hordijk, G.J., Blijham, G.H., et al. 1999. A

prospective study on quality of life of patients with cancer of the oral cavity or

oropharynx treated with surgery with or without radiotherapy. Oral Oncol. 35:27-32.

62

51. Hammerlid, E., Silander, R., Hornestam, L., Sullivan, M. 2001. Health-related quality

of life three years after diagnosis of head and neck cancer, a longitudinal study. Head

Neck. 23:113-125.

52. Gamba, A., Romano, M., Grosso, I.M., Tamburini, M., Cantu, G., et al. 1992.

Psychosocial adjustment of patients surgically treated for head and neck cancer. Head

Neck.14:218-223.

53. Netscher, D.T., Meade, R.A., Goodman, C.M., Alford, E.L., Stewart, M.G. 2000.

Quality of life and disease-specific functional status following microvascular

reconstruction for advanced (T3 and T4) oropharyngeal cancers. Plast. Reconstr. Surg.

105,1628-1634.

54. Rogers, S.N., Devine, J., Lowe, D., Shokar, P., Brown, J.S., et al. 2004. Longitudinal

health related quality of life after mandibular resection for oral cancer: A comparison

between rim and segment. Head Neck. 26:54-62.

55. Rogers, S.N., Lowe, D., McNally, D., Brown, J.S., Vaughan, E.D., 2003. Health

related quality of life after maxillectomy: A comparison between prosthetic obturation

and free flap. J. Oral. Maxillofac. Surg. 61:174-181.

63

56. Laverick, S., Lowe, D., Brown, J.S., Vaughan, E.D., Rogers, S.N. 2004. The impact

of neck dissection on health-related quality of life. Arch. Otolaryngol. Head Neck Surg.

130:149-154.

57. Licciardello, J.T., Spitz, M.R., Hong, W.K. 1989. Multiple primary cancer in patients

with cancer of the head and neck: Second cancer of the head and neck, esophagus, and

lung. Internat. J. Radiat. Oncol. Biol. Phys. 17:467-476.

58. Cohn, A.M., Peppard, S.B. 1980. Multiple primary malignant tumors of the head and

neck. Am. J. Otolaryngol. 1:411-417.

59. Shapshay, S.M., Wong, W.K., Fried, M.P., Sismanis, A., Vaughn, C.W., Strong, M.S.

1980. Simultaneous carcinomas of the esophagus and upper aerodigestive tract.

Otolaryngol. Head Neck Surg. 8:373-377.

60. Berg, J.W., Schottenfeld, D., Ritter, F. 1970. Incidence of multiple primary cancers.

Ill. Cancers of the respiratory and upper digestive system as multiple primary cancers. J.

Natl. Cancer Inst. 44:263-270.

61. Tepperman, B.S., Fitzpatrick, P.J. 1981. Second respiratory and upper digestive tract

cancers after oral cancer. Lancet. 2:547-549.

64

62. Winn, D.M., Blot, W.J. 1985. Second cancer following cancers of the buccal cavity

and pharynx in Connecticut, 1935-1982. In Multiple primary cancers in Connecticut and

Denmark. J.D. Boice, Jr., editor. Bethesda: Department of Health and Human Services.

25-46.

63. Gluckman, J.L., Crissman, J.D. 1983. Survival rates in 548 patients with multiple

neoplasms of the upper aerodigestive tract. Laryngoscope. 93:71-74.

64. Laramore, G.E., Scott, C.B., Al-Sarraf, M., Haselow, R.E., Ervin, T.J., et al. 1992.

Adjuvant chemotherapy for resectable squamous cell carcinomas of the head and neck:

report on Intergroup Study 0034. Int. J. Radiat. Oncol. Biol. Phys. 23: 706–713.

65. Bernier, J., Cooper, J.S., Pajak, T.F., van Glabbeke, M., Bourhis, J., et al. 2005.

Defining risk levels in locally advanced head and neck cancers: A comparative analysis

of concurrent postoperative radiation plus chemotherapy trials of the EORTC (#22931)

and RTOG (# 9501). Head Neck. 27:843-850.

66. Cooper, J.S., Pajak, T.F., Forastiere, A.A., Jacobs, J., Campbell, B.H., et al. 2004.

Postoperative concurrent radiotherapy and chemotherapy for high-risk squamous-cell

carcinoma of the head and neck. N. Engl. J. Med. 350:1937-1944.

65

67. Bernier, J., Domenge, C., Ozsahin, M., Matuszewska, K., Lefebvre, J.L., et al. 2004.

Postoperative irradiation with or without concomitant chemotherapy for locally advanced

head and neck cancer. N. Engl. J. Med. 350:1945-1952.

68. U.S. Preventive Services Task Force. Oral Cancer: Screening. Available from URL:

http://www.uspreventiveservicestaskforce.org/Page/Document/UpdateSummaryFinal/oral

-cancer-screening1 [accessed December 20, 2015].

69. Sakararanarayana, R., Ramadas, K., Thara, S., Muwonge, R., Thomas, G., et al. 2012.

Long term effect of visual screening on oral cancer incidence and mortality in a

randomized trial in Kerala, India. Oral Oncol. 49:314-321.

70. Elashoff, D., Zhou, H., Reiss, J., Wang, J., Xiao, H., et al. 2012. Prevalidation of

salivary biomarkers for oral cancer detection. Cancer Epidemiol. Biomarkers Prev.

21:664-72.

71. Rethman, M.P., Carpenter, W., Cohen, E.E., Epstein, J., Evans, C.A., et al. 2010.

Evidence-based clinical recommendations regarding screening for oral squamous cell

carcinomas. J. Am. Dent. Assoc. 141:509-520.

72. Ziober, A.F., Patel, K.R., Alawi, F., Gimotty, P., Weber, R.S., et al. 2006.

Identification of a gene signature for rapid screening or oral squamous cell carcinoma.

Clin. Cancer Res. 12:5960-71.

66

73. de Bree, R., Deurloo, E.E., Snow, G.B., Leemans, C.R. 2000. Screening for distant

metastases in patients with head and neck cancer. Laryngoscope. 110:397-401.

74. Dennington, M.L., Carter, D.R., Meyers, A.D. 1980. Distant metastases in head and

neck epidermoid carcinoma. Laryngoscope. 90:196-201.

75. Black, R.J., Gluckman, J.L., Shumrick, D.A. 1984. Screening for distant metastases

in head and neck cancer patients. Aust. N. Z. J. Surg. 54:527-530.

76. Cooper, J.S., Porter, K., Mallin, K., Hoffman, H.T., Weber, R.S., et al. 2009. National

Cancer Database report on cancer of the head and neck: 10‐year update. Head Neck.

31:748-758.

77. Silverman, S. Jr. 2001. Demographics and occurrence of oral and pharyngeal cancers.

The outcomes, the trends, the challenge. J. Am. Dent. Assoc. 132:7S–11S.

78. Ries, L.A.G., Hankey, B.F., Miller, B.A., Hartman, A.M., 1991. Edwards, B.K.

Cancer Statistics Review 1973–1988. Bethesda: National Cancer Institute. 91-2789.

79. Neville, B.W., Day, T.A. 2002. Oral Cancer and precancerous lesions. CA: A cancer

journal for clinicians. 52:195-215.

67

80. Rusthoven, K., Ballonoff, A., Raben, D., Chen, C. 2007. Poor prognosis in patients

with stage I and II oral tongue squamous cell carcinoma. Cancer. 112:345-51.

81. Amit, M., Yen, T.C., Liao, C.T., Chaturvedi, P., et al. 2013. Improvement in survival

of patients with oral cavity squamous cell carcinoma: An international collaborative

study. Cancer. 119:4242-4248.

82. Franceschi, D., Gupta, R., Spiro, R.H., Shah, J.P. 1993. Improved survival in the

treatment of squamous cell carcinoma of the oral tongue. Am. J. Surg. 166:360-65.

83. Liu, S.A., Wong, Y.K., Poon, C.K., Wang, C.C., Wang, C.P., et al. 2007. Risk factors

for wound infection after surgery in primary oral cavity cancer patients. Laryngoscope.

117:166-171.

84. Agra, I.M., Carvalho, A.L., Pontes, E., Campos, O.D., Ulbrich, F.S., et al. 2003.

Postoperative complications after en bloc salvage surgery for head and neck cancer. Arch.

Otolaryngol. Head Neck Surg. 2003;129:1317–1321.

85. de-Melo, G.M., Ribeiro, K.C., Kowalski, L.P., Deheinzelin, D. 2001. Risk factors for

postoperative complications in oral cancer and their prognostic implications. Arch.

Otolaryngol. Head Neck Surg. 127:828–833.

68

86. Penel, N., Fournier, C., Lefebvre, D., Lefebvre, J.L. 2005. Multivariate analysis of

risk factors for wound infection in head and neck squamous cell carcinoma surgery with

opening of mucosa. Study of 260 surgical procedures. Oral Oncol. 41:294–303.

87. Luryi, A.L., Chen, M.M., Mehra, S., Roman, S.A., Sosa, J.A., et al. 2014. Hospital re-

admission and 30-day mortality following surgery for oral cavity cancer: Analysis of

21,681 cases. Head Neck. In press.

88. Chen, M.M., Roman, S.A., Sosa, J.A., Judson, B.L. 2013. Postdischarge

Complications Predict Reoperation and Mortality after Otolaryngologic Surgery.

Otolaryngol. Head Neck Surg. 149:865-872.

89. Morris, M.S., Deierhoi, R.J., Richman, J.S., Altom, L.K., Hawn, M.T. 2014. The

Relationship Between Timing of Surgical Complications and Hospital Readmission.

JAMA Surg. 149:348-354.

90. Awad, M.I., Shuman, A.G., Montero, P.H., Palmer, F.L., Shah, J.P., et al. 2015.

Accuracy of administrative and clinical registry data in reporting postoperative

complications following surgery for oral cavity squamous cell carcinoma. Head Neck.

37:851-861.

91. Sullivan, M.C., Roman, S.A., Sosa, J.A. 2012. Does chemotherapy prior to

emergency surgery affect patient outcomes? Examination of 1912 Patients. Ann. Surg.

69

Oncol. 19:11-18.

92. Gupta, P.K., Smith, R.B., Gupta, H., Forse, R.A., Fang, X., et al. 2012. Outcomes

after thyroidectomy and parathyroidectomy. Head Neck. 34:477-484.

93. Khuri, S.F., Daley, J., Henderson, W., Hur, K., Demakis, J., et al. 1998. The

Department of Veterans Affairs' NSQIP: the first national, validated, outcome-based,

risk-adjusted, and peer-controlled program for the measurement and enhancement of the

quality of surgical care. National VA Surgical Quality Improvement Program. Ann. Surg.

228:491-504.

94. American College of Surgeons National Surgical Quality Improvement Program.

User Guide for the 2010 Participant Use Data File. June 24, 2010. Available at:

http://site.acsnsqip.org/wp-content/uploads/2012/03/2010-User-Guide_FINAL.pdf.

[Accessed June 1, 2012].

95. Shiloach, M., Frencher, S.K. Jr, Steeger, J.E., Rowell, K.S., Bartzokis, K., et

al. 2010. Toward robust information: data quality and inter-rater reliability in the

American College of Surgeons National Surgical Quality Improvement Program. J. Am.

Coll. Surg. 210:6-16.

96. American College of Surgeons. User guide for the 2013 ACS NSQIP Paticipant Use

Data File (PUF). 2014. Available from URL:

70

https://www.facs.org/~/media/files/quality%20programs/nsqip/acs_nsqip_puf_user_guide

_2013.ashx [accessed November 29, 2015].

97. American College of Surgeons. National Cancer Data Base. Available from URL:

https://www.facs.org/quality%20programs/cancer/ncdb [accessed January 4, 2014].

98. Jesup, J.M., Menck, H.R., Winchester, D.P., Hundhal, S.A., Murphy, J.P. 1996. The

National Cancer Data Base report on hospital reporting. Cancer. 78:1829-1837.

99. Raval, M.V., Bilmoria, K.Y., Stewart, A.K., Bentrem, D.J., Ko, C.Y. 2009. Using the

NCDB for cancer care improvement: an introduction to available quality assessment

tools. J. Surg. Oncol. 99:488-490.

100. Commission on Cancer. Registry Operations and Data Standards (ROADS). 1996.

Available from URL:

https://www.facs.org/~/media/files/quality%20programs/cancer/coc/roads/roads1996.ash

x [accessed June 20, 2014].

101. Commission on Cancer. Facility Oncology Registry Data Standards (FORDS). 2003.

Available from URL: http://www.facs.org/cancer/coc/fords/fordsoriginal2003.pdf

[accessed June 20, 2014].

71

102. World Health Organization. 2015. International Classification of Diseases (ICD).

Available from URL: http://www.who.int/classifications/icd/en/ (Accessed December 1,

2015).

103. American Medical Association. 2015. cpt® Code/Relative Value Search. Available

from URL: https://ocm.ama-

assn.org/OCM/CPTRelativeValueSearch.do?submitbutton=accept. [Accessed June 14,

2014].

104. American College of Surgeons National Surgical Quality Improvement Program.

User Guide for the 2010 Participant Use Data File. 2010. Available from URL:

http://site.acsnsqip.org/wp-content/uploads/2012/03/2010-User-Guide_FINAL.pdf.

[Accessed June 1, 2012].

105. Kazaure, H.S., Roman, S.A., Sosa, J.A. 2012. Association of postdischarge

complications with reoperation and mortality in general surgery. Arch. Surg. 147:1000-

1007.

106. Fritz, A., Percy, C., Jack, A., Shanmugaratnam, K., Sobin, L., et al. 2000.

International Classification of Diseases for Oncology, ICD-O. 3rd ed. Geneva: World

Health Organization, 242 pp.

72

107. Fleming, I.D., Cooper, J.S., Henson, D.E., Hutter, R.V., Kennedy, B.J., et al., editors.

1997. American Joint Committee on Cancer (AJCC) cancer staging manual. 5th edition.

Philadelphia: J. B. Lippincott. 294pp.

108. Greene FL, Page DL, Fleming ID, et al., eds. 2002. AJCC Cancer Staging

Handbook. 6th ed. New York: Springer-Verlag. 199pp.

109. American College of Surgeons. National Cancer Data Base data dictionary.

Available from URL: http://ncdbpuf.facs.org/?q=print-pdf-all [accessed June 13, 2014].

110. Stenson, K.M., Haraf, D.J., Pelzer, H., Recant, W., Kies, M.S., et al. 2000. The role

of cervical lymphadenectomy adter aggressive concomitant chemoradiotherapy: The

feasibility of selective neck dissection. Arch. Otolaryngol. Head Neck Surg. 126:950-956.

111. Mendenhall, W.M., Villaret, D.B., Amdur, R.J., Hinerman, R.W., Mancuso, A.A.

2002. Planned neck dissection after definitive radiotherapy for squamous cell carcinoma

of the head and neck. Head Neck. 24:1012-1018.

112. Mendenhall, W.M., Million, R.R., Cassisi, N.J. 1986. Squamous cell carcinoma of

the head and neck treated with radiation therapy: the role of neck dissection for clinically

positive neck nodes. Int. J. Radiat. Oncol. Biol. Phys. 12:733-740.

73

113. McGuirt, W.F., McCabe, B.F. 1980. Bilateral neck dissections. Arch. Otolaryngol.

106:427-429.

114. Brizel, D.M., Prosnitz, R.G., Hunter, S. 2004. Necessity for adjuvant neck dissection

in setting of concurrent chemoradiation for advanced head and neck cancer. Int. J. Radiat.

Oncol. Biol. Phys. 58:1418-1423.

115. Brazilian Head and Neck Cancer Study Group. 1998. Results of a prospective trial

on elective modified radical classical versus supromohyoid neck dissection in the

management of oral squamous carcinoma. Am. J. Surg.176:422-427.

116. Mosely, L.H., Finseth, F., Goody, M. 1978. Nicotine and its effect on wound

healing. Plast. Reconstr. Surg. 61:570-575.

117. Silverstein, P. 1992. Smoking and wound healing. Am. J. Med. 93:22S-24S.

118. Møller, A.M., Villebro, N., Pederson, T., Tønneson, H. 2002. Effect of preoperative

smoking intervention on postoperative complications: a randomised clinical trial. Lancet.

359:114-117.

119. Lotfi, C.J., Cavalcanti Rde, C., Costa e Silva, A.M, Latorre Mdo, R., Ribeiro Kde,

C., et al. 2008. Risk factors for surgical-site infections in head and neck cancer surgery.

Otolaryngol. Head Neck Surg. 138:74-80.

74

120. Sørensen, L.T., Hørby, J., Friis, E., Pilsgard, B., Jørgensen, T. 2002. Smoking as a

risk factor for wound healing and infection in breast cancer surgery. Euro. J. Surg. Oncol.

28:815-820.

121. Windsor, J.A., Hill, G.L. 1988. Weight loss with physiologic impairment. A basic

indicator of surgical risk. Ann. Surg. 207: 290-296.

122. Scott, H.R., McMillan, D.C., Forrest, L.M., Brown, D.J.F., McArdle, C.S., et al.

2002. The systemic inflammatory response, weight loss, performance status and survival

in patients with inoperable non-small cell lung cancer. Br. J. Cancer. 87:264-267.

123. McMillan, D.C., Watson, W.S., O'Gorman, P., Preston, T., Scott, H.R., et al. 2001.

Albumin concentrations are primarily determined by the body cell mass and the systemic

inflammatory response in cancer patients with weight loss. Nutr. Cancer. 39:210-213.

124. Schwartz, S.R., Yueh, B., Maynard, C., Daley, J., Henderson, W., et al. 2004.

Predictors of wound complications after laryngectomy: a study of over 2000 patients.

Otolaryngol. Head Neck Surg. 131: 61-68.

125. Benbassat, J., Taragin, M. 2000. Hospital readmissions as a measure of quality of

health care: advantages and limitations. Arch. Intern. Med. 160:1074-1081.

75

126. Centers for Medicare & Medicaid Services. Readmissions Reduction Program.

2014. Available from URL: http://www.cms.gov/Medicare/Medicare-Fee-for-Service-

Payment/AcuteInpatientPPS/Readmissions-Reduction-Program.html. [Accessed February

20, 2015].

127. Haughey, B.H., Wilson, E., Kluwe, L., Piccirillo, J., Frederickson, J., et al. 2001.

Free flap reconstruction of the head and neck: analysis of 241 cases. Otolaryngol. Head

Neck Surg.125:10-17.

128. Jencks, S.F., Williams, M.V., Coleman, E.A. 2009. Rehospitalizations among

patients in the Medicare fee-for-service program. N. Engl. J. Med. 360:1418-1428.

129. Tuggle, C.T., Park, L.S., Roman, S., Udelsman, R., Sosa, J.A. 2010.

Rehospitalization among elderly patients with thyroid cancer after thyroidectomy are

prevalent and costly. Ann. Surg. Oncol. 17:2816-2823.

130. Robles, L., Slogoff, M., Ladwig-Scott, E., Zank, D. 2011. The addition of a nurse

practitioner to an inpatient surgical team results in improved use of resources. Surgery.

150:711-717.

131. Pulte, D., Brenner, H. 2010. Changes in survival in head and neck cancers in the late

20th and 21st century: a period analysis. Oncologist. 15:994-1001.

76

132. Chen, M.M., Roman, S.A., Yarbrough, W.G., Burtness, B.A., Sosa, J.A., et al. 2014.

Trends and variations in the use of adjuvant therapy for patients with head and neck

cancer. Cancer. 120:3353-3360.

133. Sharma, A., Schwartz, S.M., Mendez, E. 2013. Hospital volume is associated with

survival but not multimodality therapy in Medicare patients with advanced head and neck

cancer. Cancer. 119:1845-1852.

134. Cheung, M.C., Koniaris, L.G., Perez, E.A., Molina, M.A., Goodwin, W.J., et al.

2009. Impact of hospital volume on surgical outcome for head and neck cancer. Ann.

Surg. Oncol. 16:1001-1009.

135. Eskander, A., Irish, J., Groome, P.A., Freeman, J., Gullane, P., et al. 2014. Volume-

outcome relationships for head and neck cancer surgery in a universal health care system.

Laryngoscope. 124:2081-8.

136. Wheless, S.A., McKinney, K.A., Zanation, A.M. 2010. A prospective study of the

clinical impact of a multidisciplinary head and neck tumor board. Otolaryngol. Head

Neck Surg. 143:650-654.

137. Kupersmith, J. 2005. Quality of care in teaching hospital: a literature review. Acad.

Med. 80:458-66.

77

138. Jones, A.S., Bin Hanafi, Z., Nadapalan, V., Poland, N.J., Kinsella, A., et al. 1996.

Do positive resection margins after ablative surgery for head and neck cancer adversely

affect prognosis? A study of 352 patients with recurrent carcinoma following

radiotherapy treated by salvage surgery. Br. J. Cancer. 74:128-132.

139. Joo, Y.H., Yoo, I.R., Cho, K.J., Park, J.O., Nam, I.C., et al. 2013. Standardized

uptake value and resection margin involvement predict outcomes in pN0 head and neck

cancer. Otolaryngol. Head Neck Surg. 149:721-726.

140. Luryi, A.L., Chen, M.M., Mehra, S., Roman, S.A., Sosa, J.A. 2014. Positive

Surgical Margins in Early Stage Oral Cavity Cancer An Analysis of 20,602 Cases.

Otolaryngol. Head Neck Surg. 151:984-990.

141. Bilimoria, K.Y., Stewart, A.K., Winchester, D.P., Ko, C.Y. 2008. The National

Cancer Data Base: A powerful initiative to improve cancer care in the United States. Ann.

Surg. Oncol. 15:683–690.

142. Bilimoria, K.Y., Bentrem, D.J., Stewart, A.K., Winchester, D.P., Ko, C.Y. 2009.

Comparison of Commission on Cancer–Approved and –Nonapproved Hospitals in the

United States: Implications for Studies That Use the National Cancer Data Base. J. Clin.

Oncol. 27:4177-4181.