Gait Speed and Survival in Patients With Brain Metastases

Academic Article

Abstract

  • © 2017 American Academy of Hospice and Palliative Medicine Context Accurate estimation of life expectancy in patients with brain metastases is critical for counseling and choosing appropriate therapy. Performance status is the single greatest determinant of overall survival in this population. However, current measures of performance status are subjective and often based on brief clinical encounters. Gait speed is an objective, reliable predictor of overall health and survival. Objective The purpose of this study was to evaluate the relationship between gait speed and survival in patients with brain metastases. Methods We conducted a retrospective review of all patients with documented gait speed and Karnofsky performance status seen in consultation for newly diagnosed brain metastases from 2014 to 2015. Gait speed was measured during neurological examination over 4 m at normal pace. Graded prognostic assessment scores were calculated from clinical information. The primary outcomes were overall survival and 30-day mortality. Results Eighty-five of 88 patients (97%) met inclusion criteria. Overall, the median gait speed was 0.7 m/s (range 0–1.0 m/s). Gait speed was associated with increased overall survival in addition to graded prognostic assessment score. Median survival was longer in patients with normal gait speed (>0.6 m/s, 11.9 months) compared to those with slow gait speed (≤0.6 m/s, 4.5 months, P < 0.001) or who were nonambulatory (1.1 months, P < 0.001). Thirty-day mortality for normal, slow, and nonambulatory patients was 0%, 15%, and 42%, respectively. The graded prognostic assessment overestimated actual survival for nonambulatory patients (2.2 vs. 1.1 months) and underestimated for those with normal gait speed (4.4 vs. 11.9 months). Conclusion Gait speed is associated with overall survival in patients with newly diagnosed brain metastases. Gait speed assessment is simple, objective, and may provide additional prognostic information to improve life expectancy estimation and management decisions.
  • Digital Object Identifier (doi)

    Author List

  • Dulaney CR; McDonald AM; Wallace AS; Fiveash J
  • Start Page

  • 105
  • End Page

  • 109
  • Volume

  • 54
  • Issue

  • 1