Could Gait Speed Predict Outcomes in Older Patients With Blood Cancers?

How quickly an older patient with a hematologic malignancy walks may be an indicator of that patient’s frailty and could predict for worse outcomes independent of performance status, according to the results of a new study.

“In our cohort of older adults with hematologic malignancies, gait speed predicted survival, unplanned hospitalizations, and emergency department visits, independent of cognitive status, demographic, and cancer-related risk factors,” wrote Michael Liu, MD, MS, MPH, of Harvard T. H. Chan School of Public Health, and colleagues, in the journal Blood. “It is a representative geriatric index of frailty and function that performed well even when compared to more comprehensive frailty assessments.”

The prospective study included 448 patients age 75 years or older with initial diagnosis at the myelodysplastic syndrome/leukemia, myeloma, or lymphoma clinics of the Dana-Farber Cancer Institute. Participants agreed to assessment of gait and grip. Gait speed was measured using the National Institutes of Health four-meter gait speed test. Grip strength was measured by the Jamar Hydraulic Hand Dynamometer once with each hand.

Unplanned hospital or emergency department use was documented for at least 6 months in a subset of 314 patients. Among this group, 19.1% had an unplanned hospitalization and 16.8% had an emergency department visit.

Univariate analysis showed that for every 0.1 meter per second decrease in gait speed, there was an increased hazard ratio (HR) for death (HR, 1.22; 95% CI, 1.12–1.29). Adjustment for covariates did not significantly change this result, and gait speed remained an independent predictor of mortality after including a patient’s performance status in a fully adjusted model (HR, 1.12; 95% CI, 1.01–1.24; P = .03).

Decreased gait speed was also associated with an increased odds of having an emergency department visit (odds ratio, 1.33; 95% CI, 1.10–1.61), but there was no association with unplanned hospital visits.

Grip strength was also associated with mortality. A decrease in grip strength of 5 kg was associated with a 24% increased rate of death (HR, 1.24; 95% CI, 1.07–1.43). Again, grip strength remained a significant predictor of mortality even after controlling for a patient’s performance status (HR, 1.26; 95% CI, 1.04–1.52). Decreased grip strength was not associated with emergency department or hospital use.

“The integration of such direct measures of physical function into routine clinical care could substantially improve patient assessment, prognostication, and individualization of care,” the researchers wrote.

“In other fields, gait speed is already being adopted as a ‘functional’ vital sign,” they wrote. “Our work supports the integration of gait speed into both routine clinical assessment and clinical trials of blood cancer patients, where it could serve as both an important predictor, as well as an outcome, tracking changes in function and frailty over time while on novel or existing therapies.”

Commenting on the study, Supriya G. Mohile, MD, MS, of the University of Rochester Medical Center, told Cancer Network, “Interestingly and importantly, these simple performance-based measures were comparable to other validated measures of frailty in predicting outcome. These results are clinically significant.”

“Healthcare systems that care for older patients with cancer would benefit from incorporating measures historically utilized in geriatrics to identify those risks for poor outcomes. These measures are better than standard oncology assessments such as ECOG,” she said. “Gait speed is a simple measure that could easily be integrated into routine care. Gait speed was able to select among those older adults with good performance status by ECOG, [and] those who were at higher risk of mortality.”