[Skip to Content]
[Skip to Content Landing]
Views 472
Citations 0
Original Investigation
May 2, 2019

Association of Preoperative Functional Performance With Outcomes After Surgical Treatment of Head and Neck Cancer: A Clinical Severity Staging System

Author Affiliations
  • 1Department of Otolaryngology–Head and Neck Surgery, Washington University School of Medicine in St Louis, St Louis Missouri
  • 2Statistics Editor, JAMA Otolaryngology–Head & Neck Surgery
  • 3Department of Anesthesiology, Washington University School of Medicine in St Louis, St Louis, Missouri
  • 4Editor, JAMA Otolaryngology–Head & Neck Surgery
JAMA Otolaryngol Head Neck Surg. Published online May 2, 2019. doi:10.1001/jamaoto.2019.1035
Key Points

Question  Is evaluation of preoperative functional performance, when added to other well-recognized clinical factors, associated with greater prognostic ability for outcomes following surgical treatment of head and neck cancer?

Findings  In this cohort study of 651 patients, evaluation of preoperative functional performance, comorbidity, weight loss, and TNM stage was associated with independent prognostic information for unplanned hospital readmission, complications, and overall survival. When these 4 variables were combined to create a composite clinical severity staging system, there was an association between greater severity stage and adverse outcomes.

Meaning  Poor preoperative functional performance is associated with negative outcomes for patients treated surgically for head and neck cancers, and it might be combined with other important clinically available factors to help predict adverse outcomes.


Importance  Patients with head and neck cancers have comorbidities and other constitutional symptoms known to be associated with adverse postoperative outcomes, but the role of functional performance is not well studied.

Objective  To explore the addition of functional performance to other clinical factors for association with 3 patient outcomes: 30-day unplanned readmission (UR), 90-day medical complications, and overall survival (OS).

Design, Setting, and Participants  This retrospective cohort study was conducted in a single tertiary care center with patients surgically treated for squamous cell cancer of the lip, oral cavity, pharynx, or larynx from January 2012 to December 2016. All analysis took place between January 2018 and November 2018. Data from 2 registries were analyzed, supplemented with medical record review. Logistic regression analysis was used to explore association of preoperative functional performance with outcomes. Conjunctive consolidation was used to create a useful clinical severity staging system, which included functional performance (estimated from metabolic equivalent [MET] score: <4, light-intensity activities; ≥4 at least moderate-intensity activities); overall comorbidity severity; preoperative weight loss; and TNM tumor staging. Logistic regression was used to assess the prognostic accuracy of the clinical severity staging system for 30-day UR and 90-day complications, and Cox proportional hazard regression for OS.

Exposures  All patients underwent surgical treatment for head and neck cancer.

Main Outcomes and Measures  The primary outcomes were 30-day UR and 90-day complications; the secondary outcome was OS.

Results  For the 657 patients included, the mean (SD) age was 62.0 (11.3) years; 73% were men (n = 477), and 88% were white (n = 580). A total of 75 (11%) had a 30-day UR; 204 (31%) developed a 90-day complication; and 127 (19%) patients died during the observation period. Individually, poor functional performance (<4 METs), high comorbidity burden, preoperative weight loss, and advanced TNM stage were associated with all 3 outcomes; the increased risk for each outcome ranged from 1.5 to 3.0 times the reference range. Using these 4 variables in combination, the 4-category clinical severity staging system demonstrated a strong association between severity stage and all 3 adverse outcomes: 30-day UR (C statistic, 0.63), 90-day complications (C statistic, 0.63), and OS (C statistic, 0.68).

Conclusions and Relevance  Poor preoperative functional performance, high comorbidity burden, preoperative weight loss, and advanced tumor stage were all associated with worse patient outcomes after head and neck cancer surgery. The model incorporating all 4 of these factors developed in this study may facilitate patient-centered risk assessment and patient-physician shared preoperative decision making.