A Three-Year Follow-Up On A Limb Reimplantation Performed Under Virtually Ideal Conditions

Gael R. Frank, M.D. Thelma Pedersen, R.P.T.

Articles in the lay literature have given the impression that limb reimplantation or perhaps even transplantation may someday provide the ideal prosthesis. In order to put any such optimism in its proper perspective, this paper reports a three-year follow-up on a patient whose right upper extremity was reimplanted under circumstances which approached the optimal.

Prior to this patient's arrival at the University of Oklahoma Medical Center, a surgical team consisting of orthopedic and vascular surgeons had performed preliminary experiments on dogs, reim-planting limbs which had been severed four to six hours previously. On November 2, 1964, members of this surgical team were notified that a 20-year-old workman had sustained an avulsion injury at the level of the middle third of the right humerus in a highspeed laundry centrifugal dryer. A number of circumstances helped make this an ideal case for limb reimplantation:

First, the injury was a clean-cut amputation, with the gross appearance of having been done with a very sharp cutting instrument. A short segment of the humerus had been comminuted in its midshaft portion, but other than this the injury to bone was not severe. The humerus was the only bone fractured (Fig. 1 ). Second, the injury occurred in the relatively clean environment of a laundry, and the arm, instead of being grossly contaminated with dirt, was clean when retrieved from the laundry drying machine. Third, the patient's companion, who retrieved the amputated arm from the laundry dryer, placed it in a large, clean laundry bucket with ice. Fourth, the patient was transported immediately to Oklahoma City, and arrived in our Emergency Room approximately two hours following the injury. Fifth, the surgical team had been notified prior to the arrival, and preparations were under way to reimplant the arm without undue delay. Sixth, the patient was young and well-motivated. He was also well-coordinated, as evidenced by the fact that he was a first-string player on his college basketball team.

Operative Procedures

During a six-hour operative procedure, begun approximately an hour after the patient's arrival, the right upper extremity was reimplanted. Technically no serious problems were encountered. The humerus was shortened approximately one inch through the severely comminuted portion of the fracture, and internal fixation was obtained with two plates applied to the fracture. These plates were placed at right angles to each other on the shaft of the humerus. The brachial artery and vein and the cephalic vein were sutured. Upon release of the arterial clamp, excellent back flow was obtained from the muscles and the vascular bed of the amputated extremity. The median nerve was identified and the ends were reapproximated but not repaired. The musculocutaneous nerve ends were identified and repaired. The ulna nerve ends were then identified and loosely sutured together. The radial nerve was not identified at the time of the original injury.

The wound edges healed per primam, but for about three weeks serious drainage persisted from a small sinus tract in the wound. The lymphatic channels were recannulized approximately three weeks following the injury, and the wound healed completely.

Later Nerve Repair

Approximately three months following the injury the patient was readmitted for nerve exploration and repair. It was possible to obtain end-to-end repair of the median nerve after excising approximately 12 centimeters of damaged nerve. The ends were sutured together under considerable tension, but primary repair was obtained. It was necessary to excise a 12-centimeter defect in the ulnar nerve and a portion of the medial antebrachial cutaneous nerve was removed and used as a free nerve graft to repair the defect. The radial nerve was identified in the proximal segment, but the distal segment was not identified. Hence radial nerve repair was postponed until a later date.

The patient was readmitted approximately six months following the injury for exploration of the radial nerve. At the time of this exploration it was noted that the radial nerve had been damaged over approximately 16 to 20 centimeters of its length. This defect was filled with a double sural nerve graft. X-rays taken two months following the injury showed that the fractured humerus had united (Fig. 2 ).

Muscle Function and Sensation

Physical therapy was started during the first month after surgery and has been carried out without interruption since that time. The return of muscle function and the gradual increase in the patient's use of his extremity, together with his limitations, are described below.

  1. The patient is still gaining muscle function in the extremity and has not undergone the extensive atrophy that one might expect with this type of nerve lesion. This phenomenon had been noted previously by Russian experimenters working with dogs1.

  2. We had originally felt that it might be necessary to carry out some stabilizing procedure about the wrist, but so far the dramatic return of muscle function has encouraged us, and we now plan to wait another year or two to see whether any such procedure is necessary.

  3. Although the patient perceives the sensation of pin prick in the median nerve distribution to the fingertips, he is still unable to identify objects and does not have the sense of limb presence that will be required to use the limb in his daily activities, even though the muscles are capable of performing these functions, as shown in Fig. 3 . Fig. 4 and Fig. 5

  4. He has almost completely retrained his left upper extremity and uses it exclusively in most functions.

  5. He is able to support himself on the reimplanted limb and can exercise the large muscles of the arm by doing push-ups (Fig. 6 ).

Physical Therapy

On November 9, 1964, one week after the reimplantation, the only physical therapy ordered consisted of passive exercise and electrical stimulation to the hand and wrist. At that time the entire arm was edematous, and the wrist and hand were splinted in the neutral position. Range of motion was normal in the wrist and the interphalangeal joints of the fingers; the metacarpophalangeal joints showed some slight limitation of motion. The muscles showed no response to faradic current but exhibited a fairly brisk response to galvanic. To prevent deconditioning, general body exercises were begun.

By November 17, the edema in the arm was much decreased, and passive and active assistive exercises to the elbow and shoulder were added to the treatment regimen. The shoulder could be flexed to 35 degrees and abducted to 45 degrees before the patient experienced discomfort at the site of the wound. Massage with lanolin was added to improve skin condition and to further decrease the edema.

Dimensional and Range-of-Motion Data

To study atrophy following the reimplantation, circumferential measurements of the arm (in inches) were taken periodically as follows: Table 1

On December 18, 1964, range of motion of the elbow and shoulder were also measured. The elbow range was normal; shoulder range (in degrees) was flexion, 110; abduction, 105; internal rotation, 25; and external rotation, 35.

The strength of muscles was poor. At this time whirlpool was added to the treatment armamentarium. By March 1, 1965, the range of motion in the shoulder had improved in some respects as shown by the following measurements (in degrees): flexion, 120; abduction, 105; internal rotation, 60; external rotation, 65. Range of motion of the elbow at this time was flexion, 130 degrees, and extension, 50 degrees.

By February 15, increased limitation of motion was evident in both elbow and shoulder and active exercises to the shoulder and passive and active assistive exercises to the elbow were begun. No attempt was made to forcibly increase the range of motion. At this time the possibility of sensory return to the reimplanted arm was questionable.

On April 2, 1965, the patient was evaluated for a dynamic splint with a "knuckle buster." These devices were tried, but the patient had become so proficient with his left hand that he lacked the motivation to use the dynamic splint and did not like the idea of wearing "an appliance."

By January, 1966, the range of motion of the extremity was normal with the exception of supination of the forearm and flexion of the metacarpophalangeal joints. Muscle strength was normal in the shoulder and good in the triceps and biceps. Wrist flexors were rated as fair and the extensors as poor plus. The finger flexors graded poor plus to fair plus, with some return of strength in the thumb muscles.

On the patient's return visit on May 27, 1966, sensory testing showed some return throughout the extremity, although he could not always distinguish between sharp and dull or hot and cold.

The motivation of this patient has been remarkable, and he has continued to be seen at regular intervals. Neuromuscular facilitation techniques have been added to his treatment. A night splint of Prenyl was made to hold the fingers in extension and the thumb in the functional position.

At the present time the motor return far exceeds that of the sensory return. The results of muscle tests are shown in the Table on page 14.

Lack of Sensation

It is felt that the lack of sensation is a definite hindrance to the recovery of this patient. Because of his inability, starting at midforearm, to discriminate between hard and soft and hot and cold, and to distinguish the shape of objects placed in his hand or to identify what portion of the hand is being stroked, muscle reeducation is difficult. It appears reasonable to assume that this patient could have a more functional hand if it were not for the sensory loss. However, he does use the hand more and more as a helper. This use was evidenced recently by the appearance of several paper cuts on his fingers, probably received from straightening and stacking the IBM cards he uses in his present employment.

Table 2


A three-year follow-up progress report on a case of limb reimplantation is presented. The present functional status of the limb is described and the management of the case to date is discussed. It is evident that even when reimplantation of an upper extremity is performed under virtually ideal conditions, the results are less than completely satisfactory. Failure to regain sensation in the limb has been the prime limiting factor in the case reported.

Gael R. Frank, M.D. and Thelma Pedersen, R.P.T. are associated with the University of Oklahoma Medical Center Oklahoma City, Oklahoma

1. Lapchinsky, A.G., "Recent Results of Experimental Transplantation of Preserved Limbs and Kidneys and Possible Use of This Technique in Clinical Practice," Ann NY Acad Sci, 87:539-569, 1960.