Correlation Between Visual Factors During Patient Transfer and Objective Measures of Operative Risk in Preoperative Risk Assessment

Abstract

<h3>Introduction and Objectives</h3> Frailty, an important consideration for determining a patient's surgical risk, is usually assessed informally based on clinical gestalt—the "eye-ball test." Surgeons are usually skilled at identifying high-risk patients, however, the factors underlying their unstructured assessments are unknown. The objective of our study was to explore the visual factors that may be used in an "eye-ball test" and understand how they compare to objective measures of frailty, multiple comorbidities, and operative risk. <h3>Methods</h3> A convergent mixed methods design was utilized to compare factors present in video recordings of patients transferring from a chair to exam table during vascular clinic visits to objective measures of surgical risk, including grip strength, modified frailty index-5 (mFI-5), and Fried Frailty Score. The videos were qualitatively evaluated by three independent coders. Patient transfer began in seated position in chair, followed by ambulation and transfer to the exam table, and ended in seated position on the exam table (Figure 1). Transfer length of time was measured in seconds based on time stamps. T-tests were used to compare length of transfer and Fisher's exact test was used to examine associations between visual factors and the objective measures of operative risk.Figure 1Components of a patient transfer from chair to exam table in the clinic setting. This segmentation was utilized for analysis of the video-recorded patient transfers.Figure 1 <h3>Results</h3> 30 patients and their video-taped transfers from two centers were analyzed. Mean age of participants was 71.9±9.45 years, 43.3% were female, and 18% identified with non-white race. Approximately half were categorized as frail by mFI-5 score and 50% were weak by grip strength measurement (65% convergence; P=.37). Frail patients had a longer transfer time (mean 11.6 vs 9.9 seconds; P=.19). A median of 45 coded factors based on visually present patient appearance and movement were coded. Patients who were considered frail more often used their arms to push up from sitting position (69.2% vs 25.0%; P=.04), have a medical device visible (e.g., portable oxygen, orthotic shoe, insulin pump) (38.5% vs 0%; P=.04), have a walking assistive device present (e.g., wheelchair, cane) (23.1% vs 0%; P=.22), and displayed postural instability during transfer from chair to exam table (50% vs 8.3%; P=.07). There appeared to be no difference in the presence of no-lace shoes (46.2% vs 41.7%; P=.82). <h3>Conclusions</h3> This pilot study reveals patient appearance and movement correlate with objective measures of operative risk. These visual factors may underlie unstructured frameworks utilized by vascular surgeons during preoperative risk assessment. Future work will build on this work by exploring surgeon's assessment in comparison to the visually coded factors in this study through case simulations.