Risk factor analysis and development of a prediction model for trismus after surgical treatment of buccal mucosa squamous cell carcinoma.
4 affiliations
- 1Department of Dentistry, Wuxi Stomatological Hospital, Wuxi, 214001, China.
- 2Department of Dentistry, Affiliated Hospital of Hebei University, Baoding, 071000, China.
- 3Department of Oral and Maxillofacial Head and Neck Oncology, Zhongshan Hospital, Fudan University, Shanghai, 200032, China.
- 4Department of Dentistry, Affiliated Hospital of Hebei University, Baoding, 071000, China. [email protected].
Abstract
Background
Buccal mucosa squamous cell carcinoma (BMSCC) arises in a functionally complex anatomical region adjacent to the buccinator muscle, the anterior border of the masseter muscle, the pterygomandibular raphe, the retromolar region, and the pterygomandibular space. Tumor infiltration, extended resection, post-reconstruction scarring, and postoperative radiotherapy may affect the masticatory muscle complex and mandibular movement. Compared with other oral subsites, such as tongue and floor-of-mouth cancers, postoperative trismus after BMSCC has more pronounced subsite-specific characteristics. Evidence regarding independent risk factors and risk prediction tools for postoperative trismus in this subsite remains limited. This study aimed to identify independent risk factors for trismus at 12 months after surgery for BMSCC and to develop an integrated peri-treatment risk-prediction model.Methods
We retrospectively analyzed clinical data from 312 patients who underwent surgical treatment for BMSCC at our institution between January 2020 and June 2024. The primary outcome was trismus at 12 months after surgery, defined as a maximum interincisal opening (MIO) of < 35 mm. The collected data included demographic characteristics, exposure to risk factors, preoperative assessment parameters, surgery-related variables, postoperative pathological indicators, adjuvant treatment, and follow-up information. Two investigators independently extracted data using a prespecified electronic case report form; data consistency was assessed by repeated random sampling. Univariable analyses were used to describe associations between candidate factors and the outcome; candidate predictors for the main model were prespecified according to the number of events, clinical relevance, temporal sequence, and variable overlap. A multivariable logistic regression model was developed and evaluated using receiver operating characteristic (ROC) curve analysis, calibration curves, the Hosmer-Lemeshow test, bootstrap internal validation, and decision curve analysis (DCA).Results
Among 312 patients, 98 (31.4%) developed trismus at 12 months after surgery, including 42 patients (42.9%) with mild trismus, 39 (39.8%) with moderate trismus, and 17 (17.3%) with severe trismus. Preoperative MIO was clinically comparable between the two groups (mean difference, -0.6 mm; 95% confidence interval (CI), -1.78 to 0.58; P = 0.312). Multivariable logistic regression showed that pathological depth of invasion (pDOI) ≥ 10 mm (odds ratio (OR) = 3.078, 95% CI, 1.756-5.392), subtotal or total masseter resection (OR = 4.288, 95% CI, 2.272-8.096), medial pterygoid muscle resection (OR = 2.680, 95% CI, 1.494-4.808), buccal mucosal defect area ≥ 15 cm² (OR = 2.402, 95% CI, 1.420-4.064), radiation dose ≥ 60 Gy (Gy) (OR = 2.252, 95% CI, 1.342-3.780), and masticatory muscle V50 ≥ 40% (percentage of masticatory muscle volume receiving ≥ 50 Gy; OR = 2.520, 95% CI, 1.368-4.644) were independent risk factors for postoperative trismus (all P < 0.05). The area under the curve (AUC) of the combined prediction model was 0.862 (95% CI, 0.820-0.904), and the bootstrap-corrected AUC was 0.848 (95% CI, 0.802-0.894). The Hosmer-Lemeshow test showed no evident lack of fit (χ² = 6.842, P = 0.554).Conclusions
pDOI, masseter and medial pterygoid resection status, buccal mucosal defect area, radiation dose, and irradiated masticatory muscle volume were independent risk factors for trismus after surgical treatment of BMSCC. The integrated peri-treatment prediction model based on these factors demonstrated good discrimination and acceptable calibration in internal validation in this cohort and may serve as an adjunctive reference for identifying high-risk patients and developing function-preserving strategies. Its clinical applicability requires further validation in multicenter external cohorts.
Related
- Development of a clinical prediction model for overall survival in locally advanced head and neck squamous cell carcinoma
- Development and Patient-Level Internal Validation of an MRI-Based Radiomics Machine Learning Model for Preoperative Prediction of Cervical Lymph Node Metastasis in Oral Squamous Cell Carcinoma: A Pilot Study.
- Clinicopathological Machine Learning Models for Mortality Risk Prediction in Head and Neck Squamous Cell Carcinoma: A Retrospective Analysis of the The Cancer Genome Atlas Head-Neck Squamous Cell Carcinoma (TCGA-HNSC) Cohort
- Survival and treatment selection in NUT carcinoma: a population-based study and propensity score-matched comparison with poorly differentiated head and neck squamous cell carcinoma
- Toripalimab combined with chemotherapy in the treatment of primary tracheal squamous cell carcinoma: a phase II proof-of-concept study
BibTeX
@article{sun2026risk,
title = {{Risk factor analysis and development of a prediction model for trismus after surgical treatment of buccal mucosa squamous cell carcinoma.}},
author = {Liying Sun and Wei Liu and Lei Xu and Ziqiang Ma and Boyue Li},
journal = {BMC oral health},
year = {2026},
month = jul,
doi = {10.1186/s12903-026-09352-w},
url = {https://europepmc.org/article/MED/42894189},
note = {CC BY-NC-ND},
}
RIS
TY - JOUR
TI - Risk factor analysis and development of a prediction model for trismus after surgical treatment of buccal mucosa squamous cell carcinoma.
AU - Liying Sun
AU - Wei Liu
AU - Lei Xu
AU - Ziqiang Ma
AU - Boyue Li
PY - 2026
DA - 2026/07/01
JO - BMC oral health
DO - 10.1186/s12903-026-09352-w
UR - https://europepmc.org/article/MED/42894189
AB - Background Buccal mucosa squamous cell carcinoma (BMSCC) arises in a functionally complex anatomical region adjacent to the buccinator muscle, the anterior border of the masseter muscle, the pterygomandibular raphe, the retromolar region, and the pterygomandibular space. Tumor infiltration, extended resection, post-reconstruction scarring, and postoperative radiotherapy may affect the masticatory muscle complex and mandibular movement. Compared with other oral subsites, such as tongue and floor-of-mouth cancers, postoperative trismus after BMSCC has more pronounced subsite-specific characteristics. Evidence regarding independent risk factors and risk prediction tools for postoperative trismus in this subsite remains limited. This study aimed to identify independent risk factors for trismus at 12 months after surgery for BMSCC and to develop an integrated peri-treatment risk-prediction model.Methods We retrospectively analyzed clinical data from 312 patients who underwent surgical treatment for BMSCC at our institution between January 2020 and June 2024. The primary outcome was trismus at 12 months after surgery, defined as a maximum interincisal opening (MIO) of < 35 mm. The collected data included demographic characteristics, exposure to risk factors, preoperative assessment parameters, surgery-related variables, postoperative pathological indicators, adjuvant treatment, and follow-up information. Two investigators independently extracted data using a prespecified electronic case report form; data consistency was assessed by repeated random sampling. Univariable analyses were used to describe associations between candidate factors and the outcome; candidate predictors for the main model were prespecified according to the number of events, clinical relevance, temporal sequence, and variable overlap. A multivariable logistic regression model was developed and evaluated using receiver operating characteristic (ROC) curve analysis, calibration curves, the Hosmer-Lemeshow test, bootstrap internal validation, and decision curve analysis (DCA).Results Among 312 patients, 98 (31.4%) developed trismus at 12 months after surgery, including 42 patients (42.9%) with mild trismus, 39 (39.8%) with moderate trismus, and 17 (17.3%) with severe trismus. Preoperative MIO was clinically comparable between the two groups (mean difference, -0.6 mm; 95% confidence interval (CI), -1.78 to 0.58; P = 0.312). Multivariable logistic regression showed that pathological depth of invasion (pDOI) ≥ 10 mm (odds ratio (OR) = 3.078, 95% CI, 1.756-5.392), subtotal or total masseter resection (OR = 4.288, 95% CI, 2.272-8.096), medial pterygoid muscle resection (OR = 2.680, 95% CI, 1.494-4.808), buccal mucosal defect area ≥ 15 cm² (OR = 2.402, 95% CI, 1.420-4.064), radiation dose ≥ 60 Gy (Gy) (OR = 2.252, 95% CI, 1.342-3.780), and masticatory muscle V50 ≥ 40% (percentage of masticatory muscle volume receiving ≥ 50 Gy; OR = 2.520, 95% CI, 1.368-4.644) were independent risk factors for postoperative trismus (all P < 0.05). The area under the curve (AUC) of the combined prediction model was 0.862 (95% CI, 0.820-0.904), and the bootstrap-corrected AUC was 0.848 (95% CI, 0.802-0.894). The Hosmer-Lemeshow test showed no evident lack of fit (χ² = 6.842, P = 0.554).Conclusions pDOI, masseter and medial pterygoid resection status, buccal mucosal defect area, radiation dose, and irradiated masticatory muscle volume were independent risk factors for trismus after surgical treatment of BMSCC. The integrated peri-treatment prediction model based on these factors demonstrated good discrimination and acceptable calibration in internal validation in this cohort and may serve as an adjunctive reference for identifying high-risk patients and developing function-preserving strategies. Its clinical applicability requires further validation in multicenter external cohorts.
SN - 1472-6831
ER -