Predictive Model for Optimal Enteral Feeding Tube Selection Following Oral Cavity Cancer Resection

Monday, September 30, 2024

3:24 PM – 3:30 PM EDT

Location: MBCC, 208

Disclosure(s):

Kelvin Y. Anderson, BS:

faculty for this accredited education activity has no relationship(s) with ineligible companies to disclose.

Daniel C. Adelsberg, BS:

faculty for this accredited education activity has no relationship(s) with ineligible companies to disclose.

S. Lynn Sigurdson, PhD:

faculty for this accredited education activity has no relationship(s) with ineligible companies to disclose.

William J. Magner, PhD:

faculty for this accredited education activity has no relationship(s) with ineligible companies to disclose.

Introduction: Oral squamous cell carcinoma (OSCC) resection often requires postoperative enteral feeding, typically short-term nasogastric tube (NGT) or long-term percutaneous endoscopic gastrostomy (PEG) tube. Previously, we identified multiple individual risk factors for NGT failure, leading to delayed PEG placement with increased cost and longer hospitalization. Currently, there are no formal guidelines for selecting enteral feeding method following OSCC resection. In this present study, we developed predictive models to optimize such clinical decision making.

Methods: We performed a retrospective cohort study using data from OSCC patients who underwent NGT placement following resection between 2014 and 2022 at an NCI-Designated Comprehensive Cancer Center (n = 81). Patients who proceeded to oral feeding following NGT (NGT Only group; n = 54) were compared with those that received PEG following NGT (NGT Failure group; n = 27). Logistic regression modeling was used to score and select potential predictor variables for multivariate analysis. All statistics were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC).

Results: We identified several independent factors significantly associated with NGT failure including lower BMI, female gender, current smoking, hypothyroidism, anxiety or depression, preoperative dysphagia, and clinical stage 4 OSCC. Using these factors, we generated 3 variable predictive models. Models showed that patients with both clinical stage 4 OSCC, BMI under 25, and either a history of tobacco use or preoperative difficulty swallowing were likely to fail NGT. C-statistic for models ranged from 0.769 to 0.866.

Conclusions: We generated predictive models to screen for OSCC patients with high postoperative risk of NGT failure based on preoperative patient characteristics. Our findings can be employed to optimize enteral tube feeding method selection following OSCC resection. Future studies will seek to externally validate and prospectively analyze our model to generate guidelines for enteral feeding selection in patients scheduled for OSCC resection.

Presenting Author(s)

Kelvin Y. Anderson, BS

Medical Student
Jacobs School of Medicine and Biomedical Sciences at the University at Buffalo
Buffalo, NY, United States

Disclosure(s):
faculty for this accredited education activity has no relationship(s) with ineligible companies to disclose.
Daniel C. Adelsberg, BS

Medical Student
Department of Otolaryngology, Jacobs School of Medicine and Biomedical Sciences at the University at Buffalo
Buffalo, New York, United States

Disclosure(s):
faculty for this accredited education activity has no relationship(s) with ineligible companies to disclose.

Co-Author(s)

Daniel C. Adelsberg, BS

Medical Student
Department of Otolaryngology, Jacobs School of Medicine and Biomedical Sciences at the University at Buffalo
Buffalo, New York, United States

Disclosure(s):
faculty for this accredited education activity has no relationship(s) with ineligible companies to disclose.
SB

Seth Brand, MBA

Jacobs School of Medicine and Biomedical Sciences at the University at Buffalo, United States

Disclosure information not submitted.
JG

Jae Gardella, BA

Roswell Park Comprehensive Cancer Center, USA

Disclosure information not submitted.
MP

Mary Platek, PhD

Roswell Park Comprehensive Cancer Center, United States

Disclosure information not submitted.
SS

S. Lynn Sigurdson, PhD

Research Technologist
Roswell Park Comprehensive Cancer Center
Buffalo, NY, United States

Disclosure(s):
faculty for this accredited education activity has no relationship(s) with ineligible companies to disclose.
William J. Magner, PhD

HRI Scientist, Assistant Professor of Oncology
Roswell Park Comprehensive Cancer Center
Buffalo, NY, United States

Disclosure(s):
faculty for this accredited education activity has no relationship(s) with ineligible companies to disclose.

Senior Author(s)

Ryan P. McSpadden, MD

Assistant Professor of Oncology
Roswell Park Comprehensive Cancer Center
Buffalo, NY, United States

Disclosure information not submitted.