Page 102 - Risk quantification and modification in older patients with colorectal cancer

Chapter 5
Mobility aid use was also associated with worse OS in the univariable analysis (HR 2.57, 95% CI 1.26-5.51, p=0.01; Appendix B shows the Kaplan Meier Curve) and multivariable analysis that included age, gender and tumour stage (HR 2.65, 95% CI 1.29-5.44, p=0.01). When comorbidity was included in the multivariable model for OS, the preoperative use of a mobility aid was still associated with worse OS (HR 2.62, 95% CI 1.22-5.63, p=0.013).
Low skeletal muscle mass and physical performance
Due to the relatively small group of patients, we performed subgroup analysis for OS based on the sex-specific median of skeletal muscle mass instead of sex-specific quartiles. There was a borderline significant interaction between mobility aid use and skeletal muscle mass (p=0.05) and no interaction between mobility aid use and skeletal muscle density (p=0.14). Stratified analysis showed that patients with a skeletal muscle mass below the median in combination with preoperative use of mobility aid had worse OS (HR 5.68, 95% CI 1.79-18.02, p=0.003). In patients with a skeletal muscle mass above the median, the use of a mobility aid was not associated with OS (HR 1.38, 95% CI 0.55-3.43, p=0.5).
Discussion
In a cohort of older patients (≥ 70 years of age), undergoing surgical treatment for CRC, the preoperative use of a mobility aid as derivative of “physical performance” was associated with higher morbidity and mortality. Skeletal muscle mass and density were not associated with OS, and only muscle density had weak associations with postoperative complications when the lowest and highest quartiles were compared. The importance of the assessment of physical performance when assessing skeletal muscle mass and density in older patients was further shown as patients with the lowest skeletal muscle mass in combination with preoperative use of a mobility aid had worse OS.
Our study suggests that a single radiological measurement of muscle mass and density has insufficient potential to be used for risk stratification in the majority of older colorectal patients and physical performance measures such as the use of a mobility aid, Timed Up and Go (TUG)23,24 or gait speed,9 would be of more importance. Although we did not assess the association between low
100