- Email: [email protected]
Postoperative results of opponensplasty and flexor tendon transfer in patients with spinal cord injuries
Postoperative results of opponensplasty and flexor tendon transfer in patients with spinal cord injuries Key pinch, palmar pinch, and grasp strength were evaluated after opponensplasty and flexor tendon transfer in 24 patients who had suffered cervical spinal cord injuries during a 17-year period. The patients had 57 tendon transfers: 35 opponensplasties and 22 flexor tendon transfers. The average follow-up was 4.2 years. The brachioradialis and pronator teres were the most frequently used motors for the opponensplasty and flexor tendon transfer. Key pinch strength averaged 1.47 kg (range of 0.13 to 4.70 kg). Grasp strength averaged 2.81 kg (range of trace to 10.0 kg). Palmar pinch was obtained in 45% of the extremities; the overall result was 1.04 kg (range of 0.20 to 3.00 kg). In general, patients with higher functional classifications achieved better results. We believe that intraoperative length-tension studies were an important factor in improving the results. The choice of muscle for opponensplasty or flexor transfer when two different adequate motors were available did not seem to affect the outcome of key pinch or grasp. We believe that tendon transfers are beneficial and should be considered in all patients with spinal cord injuries regardless of age at injury if at least 1 year has elapsed since injury and the patients are neurologically stable and have participated in a rehabilitation program. (J HAND SURG IOA:890-4, 1985.)
Charles M. Kelly, M.D., Alvin A. Freehafer, M.D., P. Hunter Peckham, Ph.D., and Kathy Stroh, O.T., Cleveland, Ohio
T
he care of patients with spinal cord in juries is a therapeutic challenge that should be handled with a team approach. We have been actively involved in the nonsurgical and surgical care of patients with spinal cord injuries for the past 24 years. Previous ar ticles have dealt with the methods of rehabilitation, the functional classification, and the early and late follow-up results of transfers and tenodeses in such pa tients.I-7 Other investigators have made valuable con tributions to the surgical and nonsurgical treatment of these patients. 731 Rehabilitation of the upper limb involves the com bined efforts of orthopedic surgeons, medical engi neers, and occupational therapists. This article evaluates the strengths of key pinch, pal
From the Department of Orthopedics, Case Western Reserve Uni versity, Cleveland Metropolitan General/Highland View Hospital, and Cleveland Veterans Administration Medical Center, Cleveland, Ohio. Received for publication Nov. 15, 1984; accepted in revised form Feb. 21, 1985. Reprint requests: Alvin A. Freehafer, M.D., Department of Or thopedics, 3395 Scranton Rd., Cleveland, OH 44109.
890
THE JOURNAL OF HAND SURGERY
mar pinch, and grasp in 24 patients (35 extremities) who had opponensplasties with or without flexor tendon transfers and discusses the operative and postoperative care of these patients.
Material and methods The group consisted of 24 patients (17 male and seven female). Their ages at the time of evaluation ranged from 19 to 60 years (average of 28 years). Their ages at injury ranged from 15 to 46 years (average of 23 years). The interval from the date of injury to operation ranged from 1 to 17 years (average of 2.6 years). The duration of follow-up after operation ranged from 1 to 17 years (average of 4.2 years). Of the 24 patients, 10 (13 extremities) have been previously described/ and 14 (22 extremities) were evaluated in our clinics during 1982 to 1983. These patients had a total of 57 tendon transfers: 35 oppo nensplasties and 22 flexor tendon transfers. All the limbs were grouped according to a previously described classification (Table 1).3.6 Of the 35 extrem ities, 13 were in group III, 11 were in group IV, seven were in group V, and four were in group VI. An op eration was performed no earlier than 1 year after injury
Vol. lOA , No . 6. Part 1 November 1985
and only if the patient was neurologically stable and had undergone a rehabilitation program. Follow-up data were obtained no earlier than 1 year after operation. Grasp and pinch were routinely evaluated before sur gery on all patients. The patients in group III (13 ex tremities) had either a good to fair automatic grasp (seven) or a poor or ineffective automatic grasp (six). Most of the patients in this group did not have flexor transfer. However, in four of the extremities, flexor transfers and opponensplasties were performed by shar ing a common motor-the brachioradialis (BR) muscle. Most of the patients in group IV (nine of 11 extrem ities) had an ineffective voluntary grasp; however, all had flexor transfers to enhance grasp. The patients in group V (seven extremities) had an effective voluntary grasp and did not require a flexor tendon transfer (five) or had poor to fair voluntary grasp and had this function supplemented with a flexor tendon transfer (two). All the patients in group VI (four extremities) had ineffec tive voluntary grasp supplemented by a flexor tendon transfer. Functional key pinch was not present in any patient before surgery. Patients were evaluated for strength of key pinch, palmar pinch, and grasp by means of standard pinch meters and dynamometers. Data on intraoperative length-tension studies and measured excursions of transferred muscles were avail able for 80% of the extremities operated on since 1979. Operation
The goals of operation were to provide effective key pinch and finger grasp. All opponensplasties and flexor tendon transfers were performed at one operative set ting. Seven of the extremities had posterior deltoid transfers to provide an active antagonist to the elbow flexion generated during strong BR contraction; these procedures were performed separately from the oppo nensplasty. Of the 24 patients, 11 (46%) had bilateral opponensplasties , and these also were performed at sep arate times. In the 35 opponensplasties that were performed, the transfer used was that described by Royle 28 and Thompson 30 : The insertion of the paralyzed flexor dig itorum superficialis of the ring finger was moved to the abductor pollicis brevis by using the distal edge of the flexor retinaculum as the pulley.6 The most common motors used were the BR and pronator teres (PT); the flexor carpi radialis (FCR) , extensor carpi radialis lon gus (ECRL) , flexor carpi ulnaris, and palmaris longus were used less frequently. Because wrist flexion is a valuable functional asset in patients with spinal cord injuries, the FCR should not be transferred unless no other motor is available. 6 Wrist extension is vital be
Results in patients with spinal cord injuries
891
Table I. Classification of upper extremities according to remaining voluntary function Group
Function (I) Shoulder shrug only ; (2) shoulder
II
III
IV
V
VI
control and elbow flexion Shoulder control. elbow flexion. weak BR . ECRL. and extensor carpi radialis brevis muscles Good or normal shoulder control. el bow flexion, and BR and radial wrist extensors Those of group 1lI plus functioning triceps. PT. and FCR muscles; weak finger fl exors and extensors Paralysis of only intrinsic muscles of hand Incomplete paralys is that did not fit into any other group
cause it provides tenodesis grasp; therefore, the exten sor carpi radialis brevis must have adequate strength if the ECRL is to be used to motor the opponensplasty. One muscle is usually transferred for each desired ac tion , opposition or grasp. In three patients, all of whom were in group III, sufficient motors were unavailable, and both the opponensplasties and flexor tendon trans fers were motored with the BR. Flexor tendon transfers were performed 22 times to provide voluntary grasp by transferring an available motor to the paralyzed flexor digitorum profundus ten dons. The BR was the most common motor; the PT, FCR, and ECRL were used less frequently. This pro cedure was always performed in combination with an opponensplasty. Three limbs in group V had a flexor lag of the index finger. Because the flexor digitorum profundus of the index finger was separate, it was su tured to the others , as described by Orner. 27 Muscles were transferred after soft tissue dissection to increase their range of motion (ROM) and allow for a straight-line pull. The muscles were inserted as near as possible to the length that corresponded to the peak on the intraoperatively constructed length-tension curve . The tension under which the tendon transfer was sutured was such that the length of the transferred mus cle in the neutral position of the wrist equaled the length of that muscle in the neutral position before it was cut for transfer. After completion of the transfer, a good tenodesis grasp was attained with the wrist in 45° of extension . With the wrist falling into flexion , the fingers opened adequately. Occasionally, passive wrist flexion did not bring the thumb out of the palm. In such cases , tenodesis of thumb extensors can be helpful. This pro
892
The Journal of
Kelly et al.
HAND SURGERY
Table II. Results of tendon transfer according to functional group Strength (kg) Group
Variable
Extremities, no.
III
Key pinch Grasp Palmar pinch Key pinch Grasp Palmar pinch Key pinch Grasp Palmar pinch
13 6
IV
V
Mean ± SD
0.50 0.33 0.50 2.04 5.57 1.54 2.39 2.00 0.20
11 7 5 7 3
1
Range
± 0.54 ± 0.82
0.13-2.00 T*-2.00
± 1.20 ± 2.70 ± 1.16
0.50-3.50 2.00-10.00 0.20-3.00 0.50-4.70 T*-5.00
± 1.62 ± 2.65
'T, trace.
cedure was not commonly needed-it was used in only one of the 55 opponensplasties previously reported. 6 Finally, the transfer was tested for adequate function before skin closure by electrical stimulation of the trans ferred motor. Postoperative therapy The postoperative protocol consisted of 3 months of therapy at a rehabilitation center. Because the arms of patients with spinal cord injuries also function as "legs," transfers must be able to withstand large forces, as would a lower extremity transfer. To prevent rupture of the transfer, a cast protected the extremity for 3 weeks after surgery, after which a removable splint was ap plied. Early therapy consisted of light, active ROM exer cises such as picking up nonresistive pegs or similar objects. No passive ROM exercises were done. By the fifth postoperative week, resistive exercises were grad ually started; these were increased until dismissal at 3 months. Use of wheelchairs, push-ups, and transfers were not allowed for 3 months. Patients received instructions for the activities of daily living as their functional neurologic status in creased one or more levels; new adaptive equipment for feeding or writing was also introduced. Results When the 14 patients (22 extremities) who were eval uated during 1982 to 1983 commented on their results, most said they would choose to have the operation again on a given extremity (95% of the extremities) and that they had improved function (91 % of the extremities). One patient believed that his function was unchanged, and one was dissatisfied with the muscle imbalance created by the transfers. However, neither patient had
key pinch before surgery, but did achieve effective key pinch after surgery. The overall value of key pinch for all 35 extremities was 1.47 ± 1.29 kg (mean ± SD). The resulting strength of key pinch for each functional group is shown in Table II. Before 1978, intraoperative length-tension and ten don excursion were not measured. When the values for key pinch in patients who had these studies were com pared with the results in those who did not, higher mean values were noted for patients in groups III, IV, and V who had length-tension and excursion measured. Key pinch for patients in group IV was evaluated to determine the effect of individual muscles used in the transfer. The results were similar when the most com monly used motors (that is, BR, four extremities; PT, three extremities; and ECRL, three extremities) were compared. Grasp has been measured in all patients since 1979; in this study, this value was determined in 20 extrem ities. The overall result for all groups was 2.81 ± 2.89 kg (mean ± SD) (range of trace to 10 kg). Individual results for each group are shown in Table II. Palmar pinch has been measured in all patients since 1979; nine (45%) of 20 extremities achieved palmar pinch. The overall result was 1.04 ± 1.02 kg (mean ± SD) (range of 0.20 to 3.00 kg). Individual results for each group are shown in Table II. Palmar pinch was achieved in only 17% of the extremities in group III, 71 % of those in group IV, and 33% of those in group V. Discussion Tendon transfers to improve the functional neurologic level of patients with spinal cord injuries have been used for 24 years by the senior author (A. A. F.).
Vol. lOA, No.6, Part 1 November 1985
Because of our encouraging early experiences, tendon transfers have been performed more frequently in recent years. Opponensplasties and flexor tendon transfers constitute most of the tendon transfers in the hand; tenodesis is performed only when motors are unavail able for transfer. The data were not statistically analyzed because the patients were not randomized and the extremities did not have controls. The operation was often bilateral, and the extremities were in different functional groups. Our classification system has been helpful as a guide to surgical treatment (Table III). Key pinch strength tends to increase in patients in the higher groups and has a wide range of results for each group. The improved results in the extremities that had in traoperative data available on the length-tension prop erties of the transferred muscles were encouraging. We believe they were due to (1) improved placement of the transfer in a position where it was seen to be mechan ically effective on its length-tension curve and (2) in vivo electrical stimulation of the transfer to assess its function before skin closure. The age of a patient at the time of injury did not seem to be a significant factor in final key pinch. The patients were considered for tendon transfer individ ually, and advanced age is not a reason to withhold operation. No apparent difference in key pinch was noted among the common motors (that is, BR, PT, and ECRL). We generally prefer to use the BR and PT and leave the ECRL to assist in the important function of wrist ex tension. Grasp strength was stronger in patients in group IV than in those in group III. This difference is due to the fact that most of the patients in group III relied on tenodesis for effective automatic grasp or shared a mo tor with the opponensplasty and thus had a weak vol untary grasp. Patients in groups IV and V achieved effective voluntary grasp because available motors were used for flexor tendon transfer. Grasp strength in pa tients in group V was weaker because flexor tendon transfers were not generally performed (only two of seven patients) because most patients in this group al ready had an effective voluntary grasp. In contrast, flexor tendon transfers were always performed on pa tients in group IV; thus, the resultant grasp in these patients involves a separate motor for transfer and may be stronger than the existing voluntary grasp usually present in some patients in group V. We are not ad vocating finger flexor transfers for all patients in group V because they will not only increase power but also
Results in patients with spinal cord injuries
893
Table III. Choice of surgical procedure for upper extremities according to functional classification Group
Objective
III
Key pinch
Grasp
IV, V
Key pinch Grasp
Procedure (motor for transfer) Tenodesis FPL (if no mo tor available); oppo nensplasty (BR and ECRL*) Tenodesis of finger flex ors; flexor tendon trans fer (BR and ECRL*) Opponensplasty (BR, PT, and ECRL*) Flexor tendon transfer (BR, PT, and ECRL *)
'Muscle must have adequate strength to extend wrist.
result in a significant muscle imbalance and difficulty with grasping large objects. Each case must be assessed individually with respect to the function of the opposite extremity. The choice of motor (BR or PT) did not seem to affect the result of grasp. The presence of a weak grasp does not preclude a good functional result. For example, one patient with a strong grasp in one hand and a weak grasp in the other used the strong extremity for power and the weaker one, which had strong key pinch (4.7 kg), for fine manipulation and grasping large objects. Palmar pinch to the index finger was a common oc currence, especially in group IV (71% of the extrem ities). This ability was useful for activities that were unable to be performed with key pinch alone. This was not our experience with other procedures used to attain key pinch such as flexor pollicis longus (FPL) transfer or FPL tenodesis, which yielded key pinch but no pal mar pinch. Arthrodesis of the metacarpophalangeal joint of the thumb with opponensplasty was not a common pro cedure in our patients. In 55 cases of opponensplasties previously reported, only one required arthrodesis of this joint along with the interphalangeal joint of the thumb. 6 Our policy is to perform arthrodesis of the thumb only if deformity makes it necessary. The only other indication is when the FPL is tenodesed or mo tored. Our suggested surgical procedures are time proven, technically simple, and reversible, give predictable re sults, and allow multiple-stage surgical procedures to be avoided. Patients routinely gain one or more func tional neurologic levels after surgery. Patients must be motivated and use their transfers because a muscle im
894
Kelly et al.
balance is created and deformities will occur unless the hands are used and stretched during activities of daily living. Tendon transfers improve function in patients with spinal cord injuries and allow them to become more independent and productive. These procedures may be done at any time 1 year after injury. They should be considered for all such patients regardless of sex, age, or age at injury if they have completed a rehabilitation program, are neurologically stable, have sufficient strength in motors that are expendable to arm function, and are motivated to improve their quality of life.
REFERENCES I. Freehafer AA, Mast WA: Transfer of the brachioradialis to improve wrist extension in high spinal-cord injury. J Bone Joint Surg [Am] 49 :648-52. 1967 2. Freehafer AA: Care of the hand in cervical spinal cord injuries. Paraplegia 7: 118-30. 1969 3. Freehafer AA, Von Haam E, Allen V: Tendon transfers to improve grasp after injuries of the cervical spinal cord. J Bone Joint Surg [Am] 56:951-9. 1974 4. Freehafer AA: Tendon transfers to improve grasp in pa tients with cervical spinal cord injury. Paraplegia 13: 15 21. 1975 5. Freehafer AA. Peckham PH, Keith MW: Determination of muscle-tendon unit properties during tendon transfer. J HAND SURG 4:331-9 , 1979 6. Freehafer AA, Kelly CM , Peckham PH: Tendon transfer for the restoration of upper limb function after a cervical spinal cord injury. J HAND SURG 9A:887-93, 1984 7. Weidman CD, Freehafer AA: Vocational outcome in pa tients with spinal cord injury. J Rehabil Res Dev 47:63 5, 1981 8. Bryan RS: The Moberg deltoid-triceps replacement and key-pinch operations in quadriplegia: preliminary expe riences . Hand 9:207-14, 1977 9. Curtis RM: Tendon transfers in the patient with spinal cord injury. Orthop Clin North Am 5:415-23, April 1974 10. Goldner JL: Reconstructive surgery of the hand in ce rebral palsy and spastic paralysis resulting from injury to the spinal cord. J Bone Joint Surg [Am] 37:1141-54, 1955 11 . Hentz VR, Keoshian LA : Changing perspectives in sur gical hand rehabilitation in quadraplegic patients. Plast Reconstr Surg 64:509-15, 1979 12. Hentz VR, Brown M, Keoshian LA: Upper limb recon struction in quadriplegia: functional assessment and pro posed treatment modifications. J HAND SURG 8: 119-31 , 1983 13. Henderson ED: Transfer of wrist extensors and brachio
The Journal of HAND SURGERY
radialis to restore opposition of the thumb. J Bone Joint Surg [Am] 44:513-22, 1962 14. House JH, Gwathmey FW, Lundsgaard OK: Restoration of strong grasp and lateral pinch in tetraplegia due to cervical spinal cord injury. J HAND SURG 1: 152-9, 1976 15. Lamb OW, Landry R: The hand in quadriplegia. Hand 3:31-7 , 1971 16 . Lamb OW, Landry RM: The hand in quadriplegia. Para plegia 9:204-12, 1972 17. Lamb OW, Chan KM: Surgical reconstruction of the upper limb in traumatic tetraplegia. J Bone Joint Surg [Br] 65:291-8, 1983 18 . Lipscomb PR, Elkins EC , Henderson ED: Tendon trans fers to restore function of hands in tetraplegia, especially after fracture-dislocation of the sixth cervical vertebra on the seventh. J Bone Joint Surg [Am] 40:1071-80, 1958 19. Maury M, Guillaumat M, Fran«ois N: Our experience of upper-limb transfers in cases of tetraplegia. Paraplegia 11 :245-52, 1973 20. McDowell CL, Moberg EA, Smith AG : International conference on surgical rehabilitation of the upper limb in tetraplegia. J HAND SURG 4:387-90, 1979 21 . Moberg E: Surgical treatment for absent single-hand grip and elbow extension in quadriplegia: J Bone Joint Surg [Am] 57:196-206, 1975 22. Moberg E: Reconstructive hand surgery in tetraplegia, stroke , and cerebral palsy: some basic concepts in phys iology and neurology. J HAND SURG 1:29-34, 1976 23. Moberg E: The upper limb in tetraplegia: a new approach to surgical rehabilitation . Stuttgart, 1978, Georg Thieme 24 . Moberg EA , Freehafer AA, Lamb OW, Leffert RD, Nigst H, Zancolli EA : A report from the International Federation of Societies for Surgery of the Hand: from the Committee on Spinal Cord Injuries 1980. Paraplegia 19:386-8, 1981 25 . Newman JH: The use of the key grip procedure for im proving hand function in quadriplegia. Hand 9:215-20, 1977 26. Nickel VL, Perry J, Garrett AL: Development of useful function in the severely paralyzed hand. J Bone Joint Surg [Am] 45:933-52 , 1963 27 . Orner GE Jr: Evaluation and reconstruction of the forearm and hand after acute traumatic peripheral nerve injuries. J Bone Joint Surg [Am] 50:1454-78, 1968 28 . Royle NO: An operation for paralysis of the intrinsic muscles of the thumb. JAM A 111:612-3, 1938 29 . Smith AG: Early complications of key grip hand surgery for tetraplegia. Paraplegia 19: 123-6, 1981 30. Thompson TC: A modified operation for opponens pa ralysis. J Bone Joint Surg 24:632-40, 1942 31 . Zancolli E: Surgery for the quadriplegic hand with active , strong wrist extension preserved: a study of 97 cases. Clin Orthop 112:101-13, 1975
Charles M. Kelly, M.D., Alvin A. Freehafer, M.D., P. Hunter Peckham, Ph.D., and Kathy Stroh, O.T., Cleveland, Ohio
T
he care of patients with spinal cord in juries is a therapeutic challenge that should be handled with a team approach. We have been actively involved in the nonsurgical and surgical care of patients with spinal cord injuries for the past 24 years. Previous ar ticles have dealt with the methods of rehabilitation, the functional classification, and the early and late follow-up results of transfers and tenodeses in such pa tients.I-7 Other investigators have made valuable con tributions to the surgical and nonsurgical treatment of these patients. 731 Rehabilitation of the upper limb involves the com bined efforts of orthopedic surgeons, medical engi neers, and occupational therapists. This article evaluates the strengths of key pinch, pal
From the Department of Orthopedics, Case Western Reserve Uni versity, Cleveland Metropolitan General/Highland View Hospital, and Cleveland Veterans Administration Medical Center, Cleveland, Ohio. Received for publication Nov. 15, 1984; accepted in revised form Feb. 21, 1985. Reprint requests: Alvin A. Freehafer, M.D., Department of Or thopedics, 3395 Scranton Rd., Cleveland, OH 44109.
890
THE JOURNAL OF HAND SURGERY
mar pinch, and grasp in 24 patients (35 extremities) who had opponensplasties with or without flexor tendon transfers and discusses the operative and postoperative care of these patients.
Material and methods The group consisted of 24 patients (17 male and seven female). Their ages at the time of evaluation ranged from 19 to 60 years (average of 28 years). Their ages at injury ranged from 15 to 46 years (average of 23 years). The interval from the date of injury to operation ranged from 1 to 17 years (average of 2.6 years). The duration of follow-up after operation ranged from 1 to 17 years (average of 4.2 years). Of the 24 patients, 10 (13 extremities) have been previously described/ and 14 (22 extremities) were evaluated in our clinics during 1982 to 1983. These patients had a total of 57 tendon transfers: 35 oppo nensplasties and 22 flexor tendon transfers. All the limbs were grouped according to a previously described classification (Table 1).3.6 Of the 35 extrem ities, 13 were in group III, 11 were in group IV, seven were in group V, and four were in group VI. An op eration was performed no earlier than 1 year after injury
Vol. lOA , No . 6. Part 1 November 1985
and only if the patient was neurologically stable and had undergone a rehabilitation program. Follow-up data were obtained no earlier than 1 year after operation. Grasp and pinch were routinely evaluated before sur gery on all patients. The patients in group III (13 ex tremities) had either a good to fair automatic grasp (seven) or a poor or ineffective automatic grasp (six). Most of the patients in this group did not have flexor transfer. However, in four of the extremities, flexor transfers and opponensplasties were performed by shar ing a common motor-the brachioradialis (BR) muscle. Most of the patients in group IV (nine of 11 extrem ities) had an ineffective voluntary grasp; however, all had flexor transfers to enhance grasp. The patients in group V (seven extremities) had an effective voluntary grasp and did not require a flexor tendon transfer (five) or had poor to fair voluntary grasp and had this function supplemented with a flexor tendon transfer (two). All the patients in group VI (four extremities) had ineffec tive voluntary grasp supplemented by a flexor tendon transfer. Functional key pinch was not present in any patient before surgery. Patients were evaluated for strength of key pinch, palmar pinch, and grasp by means of standard pinch meters and dynamometers. Data on intraoperative length-tension studies and measured excursions of transferred muscles were avail able for 80% of the extremities operated on since 1979. Operation
The goals of operation were to provide effective key pinch and finger grasp. All opponensplasties and flexor tendon transfers were performed at one operative set ting. Seven of the extremities had posterior deltoid transfers to provide an active antagonist to the elbow flexion generated during strong BR contraction; these procedures were performed separately from the oppo nensplasty. Of the 24 patients, 11 (46%) had bilateral opponensplasties , and these also were performed at sep arate times. In the 35 opponensplasties that were performed, the transfer used was that described by Royle 28 and Thompson 30 : The insertion of the paralyzed flexor dig itorum superficialis of the ring finger was moved to the abductor pollicis brevis by using the distal edge of the flexor retinaculum as the pulley.6 The most common motors used were the BR and pronator teres (PT); the flexor carpi radialis (FCR) , extensor carpi radialis lon gus (ECRL) , flexor carpi ulnaris, and palmaris longus were used less frequently. Because wrist flexion is a valuable functional asset in patients with spinal cord injuries, the FCR should not be transferred unless no other motor is available. 6 Wrist extension is vital be
Results in patients with spinal cord injuries
891
Table I. Classification of upper extremities according to remaining voluntary function Group
Function (I) Shoulder shrug only ; (2) shoulder
II
III
IV
V
VI
control and elbow flexion Shoulder control. elbow flexion. weak BR . ECRL. and extensor carpi radialis brevis muscles Good or normal shoulder control. el bow flexion, and BR and radial wrist extensors Those of group 1lI plus functioning triceps. PT. and FCR muscles; weak finger fl exors and extensors Paralysis of only intrinsic muscles of hand Incomplete paralys is that did not fit into any other group
cause it provides tenodesis grasp; therefore, the exten sor carpi radialis brevis must have adequate strength if the ECRL is to be used to motor the opponensplasty. One muscle is usually transferred for each desired ac tion , opposition or grasp. In three patients, all of whom were in group III, sufficient motors were unavailable, and both the opponensplasties and flexor tendon trans fers were motored with the BR. Flexor tendon transfers were performed 22 times to provide voluntary grasp by transferring an available motor to the paralyzed flexor digitorum profundus ten dons. The BR was the most common motor; the PT, FCR, and ECRL were used less frequently. This pro cedure was always performed in combination with an opponensplasty. Three limbs in group V had a flexor lag of the index finger. Because the flexor digitorum profundus of the index finger was separate, it was su tured to the others , as described by Orner. 27 Muscles were transferred after soft tissue dissection to increase their range of motion (ROM) and allow for a straight-line pull. The muscles were inserted as near as possible to the length that corresponded to the peak on the intraoperatively constructed length-tension curve . The tension under which the tendon transfer was sutured was such that the length of the transferred mus cle in the neutral position of the wrist equaled the length of that muscle in the neutral position before it was cut for transfer. After completion of the transfer, a good tenodesis grasp was attained with the wrist in 45° of extension . With the wrist falling into flexion , the fingers opened adequately. Occasionally, passive wrist flexion did not bring the thumb out of the palm. In such cases , tenodesis of thumb extensors can be helpful. This pro
892
The Journal of
Kelly et al.
HAND SURGERY
Table II. Results of tendon transfer according to functional group Strength (kg) Group
Variable
Extremities, no.
III
Key pinch Grasp Palmar pinch Key pinch Grasp Palmar pinch Key pinch Grasp Palmar pinch
13 6
IV
V
Mean ± SD
0.50 0.33 0.50 2.04 5.57 1.54 2.39 2.00 0.20
11 7 5 7 3
1
Range
± 0.54 ± 0.82
0.13-2.00 T*-2.00
± 1.20 ± 2.70 ± 1.16
0.50-3.50 2.00-10.00 0.20-3.00 0.50-4.70 T*-5.00
± 1.62 ± 2.65
'T, trace.
cedure was not commonly needed-it was used in only one of the 55 opponensplasties previously reported. 6 Finally, the transfer was tested for adequate function before skin closure by electrical stimulation of the trans ferred motor. Postoperative therapy The postoperative protocol consisted of 3 months of therapy at a rehabilitation center. Because the arms of patients with spinal cord injuries also function as "legs," transfers must be able to withstand large forces, as would a lower extremity transfer. To prevent rupture of the transfer, a cast protected the extremity for 3 weeks after surgery, after which a removable splint was ap plied. Early therapy consisted of light, active ROM exer cises such as picking up nonresistive pegs or similar objects. No passive ROM exercises were done. By the fifth postoperative week, resistive exercises were grad ually started; these were increased until dismissal at 3 months. Use of wheelchairs, push-ups, and transfers were not allowed for 3 months. Patients received instructions for the activities of daily living as their functional neurologic status in creased one or more levels; new adaptive equipment for feeding or writing was also introduced. Results When the 14 patients (22 extremities) who were eval uated during 1982 to 1983 commented on their results, most said they would choose to have the operation again on a given extremity (95% of the extremities) and that they had improved function (91 % of the extremities). One patient believed that his function was unchanged, and one was dissatisfied with the muscle imbalance created by the transfers. However, neither patient had
key pinch before surgery, but did achieve effective key pinch after surgery. The overall value of key pinch for all 35 extremities was 1.47 ± 1.29 kg (mean ± SD). The resulting strength of key pinch for each functional group is shown in Table II. Before 1978, intraoperative length-tension and ten don excursion were not measured. When the values for key pinch in patients who had these studies were com pared with the results in those who did not, higher mean values were noted for patients in groups III, IV, and V who had length-tension and excursion measured. Key pinch for patients in group IV was evaluated to determine the effect of individual muscles used in the transfer. The results were similar when the most com monly used motors (that is, BR, four extremities; PT, three extremities; and ECRL, three extremities) were compared. Grasp has been measured in all patients since 1979; in this study, this value was determined in 20 extrem ities. The overall result for all groups was 2.81 ± 2.89 kg (mean ± SD) (range of trace to 10 kg). Individual results for each group are shown in Table II. Palmar pinch has been measured in all patients since 1979; nine (45%) of 20 extremities achieved palmar pinch. The overall result was 1.04 ± 1.02 kg (mean ± SD) (range of 0.20 to 3.00 kg). Individual results for each group are shown in Table II. Palmar pinch was achieved in only 17% of the extremities in group III, 71 % of those in group IV, and 33% of those in group V. Discussion Tendon transfers to improve the functional neurologic level of patients with spinal cord injuries have been used for 24 years by the senior author (A. A. F.).
Vol. lOA, No.6, Part 1 November 1985
Because of our encouraging early experiences, tendon transfers have been performed more frequently in recent years. Opponensplasties and flexor tendon transfers constitute most of the tendon transfers in the hand; tenodesis is performed only when motors are unavail able for transfer. The data were not statistically analyzed because the patients were not randomized and the extremities did not have controls. The operation was often bilateral, and the extremities were in different functional groups. Our classification system has been helpful as a guide to surgical treatment (Table III). Key pinch strength tends to increase in patients in the higher groups and has a wide range of results for each group. The improved results in the extremities that had in traoperative data available on the length-tension prop erties of the transferred muscles were encouraging. We believe they were due to (1) improved placement of the transfer in a position where it was seen to be mechan ically effective on its length-tension curve and (2) in vivo electrical stimulation of the transfer to assess its function before skin closure. The age of a patient at the time of injury did not seem to be a significant factor in final key pinch. The patients were considered for tendon transfer individ ually, and advanced age is not a reason to withhold operation. No apparent difference in key pinch was noted among the common motors (that is, BR, PT, and ECRL). We generally prefer to use the BR and PT and leave the ECRL to assist in the important function of wrist ex tension. Grasp strength was stronger in patients in group IV than in those in group III. This difference is due to the fact that most of the patients in group III relied on tenodesis for effective automatic grasp or shared a mo tor with the opponensplasty and thus had a weak vol untary grasp. Patients in groups IV and V achieved effective voluntary grasp because available motors were used for flexor tendon transfer. Grasp strength in pa tients in group V was weaker because flexor tendon transfers were not generally performed (only two of seven patients) because most patients in this group al ready had an effective voluntary grasp. In contrast, flexor tendon transfers were always performed on pa tients in group IV; thus, the resultant grasp in these patients involves a separate motor for transfer and may be stronger than the existing voluntary grasp usually present in some patients in group V. We are not ad vocating finger flexor transfers for all patients in group V because they will not only increase power but also
Results in patients with spinal cord injuries
893
Table III. Choice of surgical procedure for upper extremities according to functional classification Group
Objective
III
Key pinch
Grasp
IV, V
Key pinch Grasp
Procedure (motor for transfer) Tenodesis FPL (if no mo tor available); oppo nensplasty (BR and ECRL*) Tenodesis of finger flex ors; flexor tendon trans fer (BR and ECRL*) Opponensplasty (BR, PT, and ECRL*) Flexor tendon transfer (BR, PT, and ECRL *)
'Muscle must have adequate strength to extend wrist.
result in a significant muscle imbalance and difficulty with grasping large objects. Each case must be assessed individually with respect to the function of the opposite extremity. The choice of motor (BR or PT) did not seem to affect the result of grasp. The presence of a weak grasp does not preclude a good functional result. For example, one patient with a strong grasp in one hand and a weak grasp in the other used the strong extremity for power and the weaker one, which had strong key pinch (4.7 kg), for fine manipulation and grasping large objects. Palmar pinch to the index finger was a common oc currence, especially in group IV (71% of the extrem ities). This ability was useful for activities that were unable to be performed with key pinch alone. This was not our experience with other procedures used to attain key pinch such as flexor pollicis longus (FPL) transfer or FPL tenodesis, which yielded key pinch but no pal mar pinch. Arthrodesis of the metacarpophalangeal joint of the thumb with opponensplasty was not a common pro cedure in our patients. In 55 cases of opponensplasties previously reported, only one required arthrodesis of this joint along with the interphalangeal joint of the thumb. 6 Our policy is to perform arthrodesis of the thumb only if deformity makes it necessary. The only other indication is when the FPL is tenodesed or mo tored. Our suggested surgical procedures are time proven, technically simple, and reversible, give predictable re sults, and allow multiple-stage surgical procedures to be avoided. Patients routinely gain one or more func tional neurologic levels after surgery. Patients must be motivated and use their transfers because a muscle im
894
Kelly et al.
balance is created and deformities will occur unless the hands are used and stretched during activities of daily living. Tendon transfers improve function in patients with spinal cord injuries and allow them to become more independent and productive. These procedures may be done at any time 1 year after injury. They should be considered for all such patients regardless of sex, age, or age at injury if they have completed a rehabilitation program, are neurologically stable, have sufficient strength in motors that are expendable to arm function, and are motivated to improve their quality of life.
REFERENCES I. Freehafer AA, Mast WA: Transfer of the brachioradialis to improve wrist extension in high spinal-cord injury. J Bone Joint Surg [Am] 49 :648-52. 1967 2. Freehafer AA: Care of the hand in cervical spinal cord injuries. Paraplegia 7: 118-30. 1969 3. Freehafer AA, Von Haam E, Allen V: Tendon transfers to improve grasp after injuries of the cervical spinal cord. J Bone Joint Surg [Am] 56:951-9. 1974 4. Freehafer AA: Tendon transfers to improve grasp in pa tients with cervical spinal cord injury. Paraplegia 13: 15 21. 1975 5. Freehafer AA. Peckham PH, Keith MW: Determination of muscle-tendon unit properties during tendon transfer. J HAND SURG 4:331-9 , 1979 6. Freehafer AA, Kelly CM , Peckham PH: Tendon transfer for the restoration of upper limb function after a cervical spinal cord injury. J HAND SURG 9A:887-93, 1984 7. Weidman CD, Freehafer AA: Vocational outcome in pa tients with spinal cord injury. J Rehabil Res Dev 47:63 5, 1981 8. Bryan RS: The Moberg deltoid-triceps replacement and key-pinch operations in quadriplegia: preliminary expe riences . Hand 9:207-14, 1977 9. Curtis RM: Tendon transfers in the patient with spinal cord injury. Orthop Clin North Am 5:415-23, April 1974 10. Goldner JL: Reconstructive surgery of the hand in ce rebral palsy and spastic paralysis resulting from injury to the spinal cord. J Bone Joint Surg [Am] 37:1141-54, 1955 11 . Hentz VR, Keoshian LA : Changing perspectives in sur gical hand rehabilitation in quadraplegic patients. Plast Reconstr Surg 64:509-15, 1979 12. Hentz VR, Brown M, Keoshian LA: Upper limb recon struction in quadriplegia: functional assessment and pro posed treatment modifications. J HAND SURG 8: 119-31 , 1983 13. Henderson ED: Transfer of wrist extensors and brachio
The Journal of HAND SURGERY
radialis to restore opposition of the thumb. J Bone Joint Surg [Am] 44:513-22, 1962 14. House JH, Gwathmey FW, Lundsgaard OK: Restoration of strong grasp and lateral pinch in tetraplegia due to cervical spinal cord injury. J HAND SURG 1: 152-9, 1976 15. Lamb OW, Landry R: The hand in quadriplegia. Hand 3:31-7 , 1971 16 . Lamb OW, Landry RM: The hand in quadriplegia. Para plegia 9:204-12, 1972 17. Lamb OW, Chan KM: Surgical reconstruction of the upper limb in traumatic tetraplegia. J Bone Joint Surg [Br] 65:291-8, 1983 18 . Lipscomb PR, Elkins EC , Henderson ED: Tendon trans fers to restore function of hands in tetraplegia, especially after fracture-dislocation of the sixth cervical vertebra on the seventh. J Bone Joint Surg [Am] 40:1071-80, 1958 19. Maury M, Guillaumat M, Fran«ois N: Our experience of upper-limb transfers in cases of tetraplegia. Paraplegia 11 :245-52, 1973 20. McDowell CL, Moberg EA, Smith AG : International conference on surgical rehabilitation of the upper limb in tetraplegia. J HAND SURG 4:387-90, 1979 21 . Moberg E: Surgical treatment for absent single-hand grip and elbow extension in quadriplegia: J Bone Joint Surg [Am] 57:196-206, 1975 22. Moberg E: Reconstructive hand surgery in tetraplegia, stroke , and cerebral palsy: some basic concepts in phys iology and neurology. J HAND SURG 1:29-34, 1976 23. Moberg E: The upper limb in tetraplegia: a new approach to surgical rehabilitation . Stuttgart, 1978, Georg Thieme 24 . Moberg EA , Freehafer AA, Lamb OW, Leffert RD, Nigst H, Zancolli EA : A report from the International Federation of Societies for Surgery of the Hand: from the Committee on Spinal Cord Injuries 1980. Paraplegia 19:386-8, 1981 25 . Newman JH: The use of the key grip procedure for im proving hand function in quadriplegia. Hand 9:215-20, 1977 26. Nickel VL, Perry J, Garrett AL: Development of useful function in the severely paralyzed hand. J Bone Joint Surg [Am] 45:933-52 , 1963 27 . Orner GE Jr: Evaluation and reconstruction of the forearm and hand after acute traumatic peripheral nerve injuries. J Bone Joint Surg [Am] 50:1454-78, 1968 28 . Royle NO: An operation for paralysis of the intrinsic muscles of the thumb. JAM A 111:612-3, 1938 29 . Smith AG: Early complications of key grip hand surgery for tetraplegia. Paraplegia 19: 123-6, 1981 30. Thompson TC: A modified operation for opponens pa ralysis. J Bone Joint Surg 24:632-40, 1942 31 . Zancolli E: Surgery for the quadriplegic hand with active , strong wrist extension preserved: a study of 97 cases. Clin Orthop 112:101-13, 1975