Phantom Limb Sensory Feedback Through Nerve Transfer Surgery

Despite numerous attempts to provide closed loop control (1), commercial prostheses are essentially open-loop devices that provide little or no feedback to the amputee. Indirect pressure feedback has been attempted by methods including vibration (2) and functional electrical stimulation (3, 4). These forms of feedback require conversion from pressure to another sensation (vibration or electrical stimulation). As a result, while providing information to the user, it is likely that it comes at the cost of increased mental load and low levels of information transfer (5). Patterson and Katz (6) have obtained better qualitative feedback with pressure to pressure feedback than with pressure to vibratory or electrical stimulation feedback, offering support for this idea. An adaptive process is still involved since the subject must learn to associate pressure sensed in one area with pressure felt in another area. Ideally, the sensory nerve endings of the amputated area would be stimulated in direct correlation to the function of the prosthesis. The recent work of Kuiken (7) has made this concept feasible by the use of targeted hyper reinnervation, in which nerves that innervated the arm have been transferred to chest muscle. As a result the potential exists for the subject to feel as if touch, pressure and even hot or cold temperatures are being exerted on their phantom hand. This study examines the potential of this intuitive pressure feedback.

Using targeted hyper-reinnervation to transfer nerves from a lost limb to denervated muscle as shown in Figure 1a, sensation of the lost limb was achieved on the chest of a subject (8). Four independently controlled nerve-muscle units were created by surgically anastomosing residual brachial plexus nerves to dissected and divided aspects of the pectoralis major and minor muscles. Sensory reinnervation also occurred on the chest in an area where the subcutaneous fat was removed.

As a result of this surgery, the subject perceived touch, sharp/dull and temperature sensation that he felt in his phantom limb when pin pricks or thermal changes were applied to the chest, as shown in Figure 1b. A representation was acknowledged: pushing in one area elicited perceived pressure in the palm of the hand, in another area on the back of the hand, and so on. In some areas the patient had low sensory thresholds (2 g/mm2) that he felt in his phantom arm. In other areas, while the subject perceived light touch on his chest, with greater pressure he only felt sensation in his phantom arm. We believe in these cases that the skin was not reinnervated, but added pressure stimulated nerves directly under the skin.

We have developed a motor to press on the subject’s skin in these areas. In preliminary testing the motor was located in area 1 of Figure 1b. However, the close proximity of the motor to EMG sensors prevented the subject from perceiving pressure gradation when he tensed his muscles to move the prosthetic arm. The motor was moved to a more lateral area adjacent to areas 2 and 5 in Figure 1b that corresponded to a localized area the size of a pen cap between the 4th and 5th metacarpals of the subject. This allowed the subject to perceive pressure gradation while actuating his prosthesis.

The subject was asked to discriminate between a series of pressure ranges. For each range, the subject was asked to determine the lower of two pressures for eight trials. If the subject correctly guessed 7 of those 8 trials, it was concluded that the subject could accurately discriminate that magnitude of force difference. This same test was done while the subject’s EMG exertion determined the feedback pressure. A threshold would appear on the screen, and the subject would be required to exert enough force on his terminal device to reach that level of pressure. After relaxing his grip, another force level would be shown, and the subject would then have to determine which force level had been higher.

Slightly above the threshold of perception, at 1.7 g/mm² pressure, the subject could discriminate 0.65 g/mm². In the middle of the force range at 4.4 g/mm² pressure, the subject’s pressure discrimination improved to 0.48 g/mm². Near the discomfort threshold at 6.6 g/mm² pressure, the subject could discriminate 0.52 g/mm². When the subject controlled the force, at 3.6 g/mm² pressure, the subject could discriminate 1.5 g/mm².

Initial results indicate that adequate sensory feedback exists to provide physiologically appropriate feedback. In future designs, the subject would appreciate less pinpoint accuracy feedback to his perceived phantom limb. In order to achieve this with pressure gradation (ie, not hitting a nerve), it may be necessary to pinpoint several areas on his chest in order to give a more global reading on his phantom hand. Future experiments will take advantage of improved functional characteristics of the terminal device in an attempt to objectively evaluate potential advantages of this novel form of phantom limb feedback.

This work was supported in part by a National Defense Science and Engineering Graduate Fellowship, the Department of Veterans Affairs, Rehabilitation Research and Development Service administered through the Jesse Brown VA Medical Center, Chicago, and the National Institute of Disability and Rehabilitation Research of the United States Department of Education under grant H133E980023. The opinions in this paper are those of the authors and do not necessarily reflect those of the Department of Education.

Figure 1: Somatic Representation of Nerve Transfer
a) Diagram of nerve-muscle graft procedure
b) Diagram of sensory reinnervation of anterior chest wall indication where
touching the skin surface produced sensation in his phantom arm.

Figure 2: Feedback motor
a) Motor pushes against subject’s chest
b) New version of the motor reduces profile.

References

1. D. S. Childress, "Closed-Loop Control in Prosthetic Systems - Historical Perspective," Annals of biomedical engineering, vol. 8, pp. 293-303, 1980.

2. G. F. Shannon, "A comparison of alternative means of providing sensory feedback on upper limb prostheses," Medical & biological engineering & computing, vol. 14, pp. 289- 294, 1976.

3. P. Nohama, A. V. Lopes, and A. Cliquet, "Electrotactile Stimulator for Artificial Proprioception," Artificial organs, vol. 19, pp. 225-230, 1995.

4. G. Wang, X. Zhang, J. Zhang, and W. A. Gruver, "Gripping force sensory feedback for a myoelectrically controlled forearm prosthesis," presented at IEEE International Conference on Intelligent Systems for the 21st Century, Vancouver, BC, 1995.

5. C. A. Phillips, "Sensory Feedback-Control of Upper-Extremity and Lower-Extremity Motor Prostheses," Critical reviews in biomedical engineering, vol. 16, pp. 105-140, 1988.

6. P. E. Patterson and J. A. Katz, "Design and Evaluation of a Sensory Feedback-System that Provides Grasping Pressure in a Myoelectric Hand," Bulletin of prosthetics research, vol. 29, pp. 1-8, 1992.

7. T. A. Kuiken, "Consideration of nerve-muscle grafts to improve the control of artificial arms," Journal of Technology and Disability, vol. 15, pp. 105-111, 2003.

8. T. A. Kuiken, G. A. Dumanian, R. D. Lipschutz, L. A. Miller, and K. A. Stubblefield, "The use of targeted muscle reinnervation for improved myoelectric prosthesis control in a bilateral shoulder disarticulation amputee.," Prosthetics and Orthotics International, vol. 28, pp. 245-253, 2004.