Distinctive Steady-State Heart Rate and Blood Pressure Responses to Passive Robotic Leg Exercise and Functional Electrical Stimulation during Head-Up Tilt

Abstract

Introduction: Tilt tables enable early mobilization of patients by providing verticalization. But there is a high risk of orthostatic hypotension provoked by verticalization, especially after neurological diseases such as spinal cord injury. Robot-assisted tilt tables might be an alternative as they add passive robotic leg exercise (PE) that can be enhanced with functional electrical stimulation (FES) to the verticalization, thus reducing the risk of orthostatic hypotension. We hypothesized that the influence of PE on the cardiovascular system during verticalization (i.e., head-up tilt) depends on the verticalization angle, and FES strengthens the PE influence. To test our hypotheses, we investigated the PE effects on the cardiovascular parameters heart rate (HR), and systolic and diastolic blood pressures (sBP, dBP) at different angles of verticalization in a healthy population. Methods: Ten healthy subjects on a robot-assisted tilt table underwent four different study protocols while HR, sBP, and dBP were measured: (1) head-up tilt to 60° and 71° without PE; (2) PE at 20°, 40°, and 60° of head-up tilt; (3) PE while constant FES intensity was applied to the leg muscles, at 20°, 40°, and 60° of head-up tilt; (4) PE with variation of the applied FES intensity at 0°, 20°, 40°, and 60° of head-up tilt. Linear mixed models were used to model changes in HR, sBP, and dBP responses. Results: The models show that: (1) head-up tilt alone resulted in statistically significant increases in HR and dBP, but no change in sBP. (2) PE during head-up tilt resulted in statistically significant changes in HR, sBP, and dBP, but not at each angle and not always in the same direction (i.e., increase or decrease of cardiovascular parameters). Neither adding (3) FES at constant intensity to PE nor (4) variation of FES intensity during PE had any statistically significant effects on the cardiovascular parameters. Conclusion: The effect of PE on the cardiovascular system during head-up tilt is strongly dependent on the verticalization angle. Therefore, we conclude that orthostatic hypotension cannot be prevented by PE alone, but that the preventive effect depends on the verticalization angle of the robot-assisted tilt table. FES (independent of intensity) is not an important contributing factor to the PE effect.

Keywords: cardiovascular system; functional electrical stimulation (FES); linear mixed models; orthostatic hypotension; parametric bootstrap; rehabilitation robotics; robotic tilt table.

Figure 1

Erigo tilt table: verticalization is provided by changing the inclination angle of the tilt table α. Passive Robotic leg exercise is provided through a leg drive with an adjustable speed f_step. The table is further enhanced with electrical stimulation module which enables providing electrical stimulation to the leg muscles with adjustable parameters (here current I_FES) during robotic leg exercise. Picture is copyrighted by Hocoma AG, Switzerland, and is adapted with permission.

Figure 2

(A) Study protocol 1: head-up tilt to 60° and 71° (two experiments). (B) Study protocols 2 and 3: PE at 48 steps/min (solid) without or with application of the minimum FES amplitude (dashed). The protocols were conducted at α = {20°, 40°, 60°}. The figure shows the experiment performed at each specific tilt angle. (C) Study protocol 4: during PE at 48 steps/min (solid) with FES, the FES amplitude (dashed) was changed to a higher level, i.e., 0.8I_MAX. The protocol was conducted at α = {0°, 20°, 40°, 60°}. The figure shows the experiment performed at each specific tilt angle. The highlighted areas T₁ and T₂ show the data range used for the analysis.

Figure 3

Statistical models for ΔHR, ΔsBP, and ΔdBP responses (changes with respect to supine position) to head-up tilt alone (A–C). The steady-state values for each subject are connected with a line. The highlighted areas show 95% CI. The signs ^*, ^**, and ^*** mark significant findings with p ≤ 0.05, 0.01, and 0.001, respectively. n.s. marks non-significant differences.

Figure 4

Statistical models for ΔHR, ΔsBP, and ΔdBP responses to PE (independent of FES) during head-up tilt (A–C). Steady-state values correspond to the average response of two conditions, i.e., PE without and with FES application, and for each subject they are connected with lines. The highlighted areas show 95% CI. The signs ^* and ^*** mark significant findings with p ≤ 0.05 and 0.001, respectively. n.s. marks non-significant differences.