ABSTRACT
The COVID-19 pandemic has limited liver transplant (LT) candidates access to clinics. Telehealth methods to assess frailty are needed. We developed a method to estimate the step length of LT candidates, which would permit remotely obtaining the 6-min walk test (6MWT) distance with a personal activity tracker (PAT). Methods: 6MWT was performed while candidates wore a PAT. On first 21 subjects (stride cohort), the step length was measured and compared with calculated one (6MWT-distance/6MWT steps). On a second cohort (PAT-6MWT; n = 116), we collected the 6MWT step count and used multivariable models to generate formulas estimating step length. We multiplied the estimated step length times 6MWT steps to estimate the distance and compared it to the measured distance. The liver frailty index (LFI) and 6MWT were used as frailty metrics. Results: Measured/calculated step length were highly correlated (ρ = 0.85; P < 0.001) in the stride cohort. In the PAT-6MWT cohort, LFI was the strongest variable associated with step length, along with height, albumin, and large-volume paracentesis (R 2 = 0.58). On a second model without LFI, age, height, albumin, hemoglobin, and large-volume paracentesis were strongly associated with step length (R 2 = 0.45). There was a robust correlation between observed 6MWT and PAT-6MWT utilizing step length equations with (ρ = 0.80; P < 0.001) or without LFI (ρ = 0.75; P < 0.001). Frailty by 6MWT <250 m did not change significantly using the observed (16%) or the with/without LFI-estimated (14%/12%) methods. Conclusions: We created a method to obtain 6MWT distance remotely with the use of a PAT. This novel approach opens the possibility of performing telemedicine PAT-6MWT to monitor LT candidates' frailty status.