Feasibility of using head and neck CT imaging to assess skeletal muscle mass in head and neck cancer patients

Justin E Swartz, Ajit J Pothen, Inge Wegner, Ernst J Smid, Karin M A Swart, Remco de Bree, Loek P H Leenen, Wilko Grolman

Research output: Contribution to journalArticleAcademicpeer-review


OBJECTIVES: Patients with head and neck cancer (HNC) have a higher risk of malnutrition and sarcopenia, which is associated with adverse clinical outcome. As abdominal CT-imaging is often used to detect sarcopenia, such scans are rarely available in HNC patients, possibly explaining why no studies investigate the effect of sarcopenia in this population. We correlated skeletal muscle mass assessed on head and neck CT-scans with abdominal CT-imaging.

METHODS: Head and neck, and abdominal CT-scans of trauma (n=51) and HNC-patients (n=52) were retrospectively analyzed. On the head and neck CT-scans, the paravertebral and sternocleidomastoid muscles were delineated. On the abdominal CT-scans, all muscles were delineated. Cross-sectional area (CSA) of the muscles at the level of the C3 vertebra was compared to CSA at the L3 level using linear regression. A multivariate linear regression model was established.

RESULTS: HNC-patients had significantly lower muscle CSA than trauma patients (37.9 vs. 45.1cm2, p<0.001, corrected for sex and age). C3 muscle CSA strongly predicted L3 muscle CSA (r=0.785, p<0.001). This correlation was stronger in a multivariate model including sex, age and weight (r=0.891, p<0.001).

DISCUSSION: Assessment of skeletal muscle mass on head and neck CT-scans is feasible and may be an alternative to abdominal CT-imaging. This method allows assessment of sarcopenia using routinely performed scans without additional imaging or additional patient burden. Identifying sarcopenic patients may help in treatment selection, or to select HNC patients for physiotherapeutic or nutritional interventions to improve their outcome.

Original languageEnglish
Pages (from-to)28-33
Number of pages6
JournalOral Oncology
Publication statusPublished - Nov 2016
Externally publishedYes

Cite this