|Home | About | Journals | Submit | Contact Us | Français|
In severe obstetrical brachial plexus palsy with proximal nerve root involvement, there is an insufficient number of motor axons to reconstruct the entire plexus, and neurotization procedures are the only possibility to achieve useful upper extremity function. One of the most useful neurotization procedures is intercostal nerve transfer. In our practice, intercostal nerve transfer was used for direct neurotization of primary nerve targets or for neurotization of transferred muscles. The best results were seen after intercostal nerve transfer to musculocutaneous nerve and triceps branch. Unlike adult posttraumatic brachial plexus patients, neurotization of the ulnar and median nerves in obstetrical brachial plexus palsy patients will result in protective sensation in almost all patients, and if used early enough, postinjury active wrist and finger flexion is possible.
In obstetrical brachial plexus palsy, when the proximal nerve root stumps are involved and are not sufficient to ensure an adequate number of motor axons to reconstruct the entire plexus, neurotization procedures are the only possibility to achieve useful upper extremity function. The neurotization procedures for brachial plexus reconstruction can be achieved by using intraplexus motor donors (the proximal nerve stump must be available) or extraplexus donors such as intercostal nerves (ICNs), cervical plexus motors, distal spinal accessory nerve, dorsal scapular nerve, and hypoglossal nerve.1,2,3,4,5,6
Neurotization procedures have also been used in late obstetrical brachial plexus cases when free muscle transfer to the upper extremity is indicated.7,8,9 ICN transfer will provide a good source of axons for the nerve of the distal target.10,11,12 During this procedure, usually two or more ICNs are separated from their territory, and their proximal stump is connected directly or via graft to a distal nonfunctioning nerve or coapted with the nerve of a transferred muscle. Thus, a given peripheral function now will be controlled by medullary or cerebral centers that before performed other functions.11,12
Yeoman and Seddon,13,14 in 1961, at the Royal National Orthopaedic Hospital in London, transferred several ICNs via ulnar nerve graft into the musculocutaneous nerve, reporting a good result. This technique was applied and further developed by several investigators and is commonly used for brachial plexus reconstruction.10,15,16,17,18,19,20,21,22 The experience with ICN neurotization is well documented and prescribed in several reports for adult posttraumatic brachial plexus patients, but to date, few papers have been written on the results of using ICN transfer in obstetrical brachial plexus palsy (OBPP) patients.
ICNs are the ventral primary rami of spinal nerves T1–T11. For brachial plexus reconstruction purposes, nine ICNs are available: T3–T11. The nerve T1 is a component of the brachial plexus, whereas the T2 ICN is not accessible for harvesting. The nipple area is usually supplied by the anterior sensory ramus of the fourth ICN, and its use should be avoided. An ICN, along with the intercostal vessels, runs parallel and under its corresponding rib, in a plane between the middle and innermost intercostal muscles. The lower ICNs (T7–T11) pass behind the costal cartilages between the obliqus internus and transversus abdominis to the sheath of the rectus abdominis, which they perforate. They supply the rectus abdominis muscle and will end as the anterior cutaneous branches, supplying the skin on the front of the abdomen. Lateral cutaneous branches, after they pierce the external intercostal and serratus anterior muscles, will be divided into anterior and posterior branches. The lateral cutaneous branch of the second ICN is named the intercostobrachial nerve and does not divide like the others into anterior and posterior branches, but will cross the axilla to the medial side of the arm and joins with the medial brachial cutaneous nerve. This nerve supplies the skin of the upper half of the medial and posterior part of the arm, communicating with the posterior brachial cutaneous branch of the radial nerve.23
Preoperative evaluation included a thorough maternal and birth history. The presence of complications during the pregnancy, the weeks of gestation, Apgar scores, perinatal complications such as shoulder dystocia, and possible clavicle or humerus fractures were recorded and documented in the patient chart. Preoperative evaluation also included a thorough manual testing of each muscle on the affected side, always comparing it with the normal side, as well as careful assessment for active and passive range of motion, especially on the elbow and shoulder joints; videotape documentation; and plain x-rays to rule out the coexistence of clavicle or humeral fractures. Inspiratory/expiratory chest x-rays and chest fluoroscopy to evaluate the phrenic nerve function. A paralyzed diaphragm, determined by the inspiratory/expiratory chest x-ray and chest fluoroscopy, indicates a high plexus involvement and is a contraindication for neurotization via ICNs. Computed tomography myelography and plain myelography will demonstrate the presence or absence of the pseudomeningoceles on the cervical spine.5 Although difficult to perform in the neonate, a sensory evaluation—the reaction to stimuli from light touch to pinprick, cutaneous pressure thresholds with Semmes-Weinstein monofilament, the evidence of sensory disturbances such as trophic changes, and the presence of a cool and dry skin is described in each child. A ninhydrin sweat test is always performed during the preoperative and postoperative follow-up evaluations to estimate the sudomotor function. Along with the above-mentioned tests, the presence and advance of the Tinel sign is checked during each follow up, photographed, and recorded in the patient's chart.
Some infants with multiple root avulsion may show signs of “amblyopia” by ignoring the affected side and may develop torticollis by turning the head on the opposite side. The presence of a Horner sign implies a severe injury and is indicative of avulsion at C8–T1 roots. An angiogram of the affected upper extremity is performed preoperatively in cases of free muscle transfer. Although electrodiagnostic study is very difficult in babies because of a violent reaction to the needle introduction and stimulation, this remains one of the most valid techniques of assessing functional recovery in brachial plexus lesions. The electromyogram and nerve conduction velocity study is performed routinely during the preoperative evaluation of the child to assist with the diagnosis and the reconstructive plan, as well as during each postoperative follow-up evaluation to estimate the return of function on the reinnervated muscles.
Functional recovery is assessed on the basis of the Mallet score for shoulder abduction and elbow flexion–extension function, and the Gilbert-Raimondi scale is used to estimate hand function restoration.24,25 The strength of the reinnervated muscles is graded as follows: a grade of M0 to M2 is a poor result, M2+ to M3 is fair, M3+ to M4− is good, and M4 to M5- is excellent.10
A curved incision starts at the midaxillary line to the lower chest and extends toward the midline of the abdomen at the level of the umbilicus. The extent of this incision depends on the number of ICNs to be harvested. The pectoralis major and minor muscles are reflected upward without being detached from their origin. The ICNs can be exposed by elevating the periosteum of the corresponding ribs or by dissecting the intercostal muscles. The lateral cutaneous branch is identified at the level of the anterior axillary line, close to the border of the serratus anterior with the external oblique muscle. Following the lateral cutaneous branch proximally by splitting the external and internal intercostal muscles, the motor branch is seen under the inferior surface of the rib on top of the innermost intercostal muscle and is immediately stimulated to confirm its identity. The motor ICNs are dissected far anteriorly, at the level of the costochondral junction, and posteriorly to the posterior axillary line. Care should be taken not to cause any injury to the long thoracic nerve while dissecting the motor branch of ICN proximally. In addition, care should be taken not to enter the pleura and cause pneumothorax. To achieve target connectivity, the ICNs are mobilized as far as possible. For this reason the lower ICNs are dissected under the costal cartilage, followed underneath the rectus abdominis, and cut when they penetrate the muscle ~1 cm from the linea alba. The ICNs are passed through a subcutaneous tunnel to the axilla and coapted, usually directly, with the target nerves without tension, using 10–0 nylon.
In our patient population, the ICNs were used for reconstruction of the musculocutaneous nerve, axillary nerve, triceps branch, and ulnar or median nerves.26 Furthermore, ICNs have been used for long thoracic nerve neurotization to provide a stable scapula, and for neurotization of the thoracodorsal nerve for the latissimus dorsi muscle to be used in the future as a pedicle muscle flap for enhancing elbow flexion or extension. ICNs also have been used for neurotization of free muscles in late obstetrical brachial plexus patients for elbow and hand reanimation.
An ICN contains no more than 1200–1300 myelinated fibers, and only 40% of them are motor fibers.23 Axillary and musculocutaneous nerve contains ~6000 myelinated fibers.27 Therefore, to match the requirements of the recipient nerve with donor nerve, at least two or three ICNs have to be transferred. In addition, during the neurotization procedure there is a risk of wasting an important proportion of the transferred motor nerve fibers into improper channels. For this reason, the distal site of coaptation must be as close as possible to the point at which motor fascicles entering the muscle are already identified.
Intercostal transfer to the musculocutaneous nerve (MCN) for elbow flexion restoration is one of the most commonly used neurotization procedures.28,29,30,31,32 In our obstetrical series, 15 ICNs were used to directly neurotize the musculocutaneous nerve in six patients. In four patients, three ICNs were used, and in two patients, one ICN was used for MCN neurotization in end-to-side fashion to enhance preexisting but weak muscle power. Kawabata33 reported his experience with this procedure in OBPP patients and found that, overall, 84% of his patients achieved a muscle grading of M4, whereas those patients who were operated on earlier than 5 months of age achieved this result. Malessy34 reviewed the ICN to MCN transfers in adult patients that were reported in the literature by six different centers and found that a muscle grading of M3 or better was achieved in 78% of the patients. Nagano35 reported that in his series, 70% of children achieved a muscle grading M4 after ICN to MCN transfer. Krakauer and Wood31 showed that six of eight patients achieved a muscle grading M3 or more after MCN neurotization with the ICNs, whereas Millesi36 reported M3 or better elbow flexion in more than 40% of the patients in his series. In our series, postoperatively five out of six patients achieved an excellent result (muscle grading M4 to M5−), Fig. Fig.1.1. The initial signs of recovery after ICN to MCN neurotization procedure were seen at 4–6 months, and maximal recovery was achieved 14–20 months after the surgery. With MCN neurotization from ICNs, the elbow flexion initially was found to depend on the respiration, and the patient usually experienced involuntary elbow flexion while sneezing or coughing. This involuntary contraction with breathing remains present for ~16–24 months and then, in most of the cases, becomes voluntary.32,35
Doi37,38 reported that the functional outcomes after ICN transfer to the triceps in his patients were poor. Six of 19 patients (32%) in his series achieved a muscle grading of M3 or more. In our series, a total of 11 patients underwent intercostal to triceps nerve neurotization to restore elbow extension function. To maintain the preexisting muscle function and to not downgrade, in two patients the repair was done in an end-to-side fashion, whereas in the rest of the patients the nerve repair was done in an end-to-end fashion. Five of 11 patients (46%) achieved an excellent result (muscle grading M4 to M5−) (Fig. 2).
Five children in our series had direct neurotization of ICNs to axillary nerve. In addition, all patients in this group underwent a distal spinal accessory to suprascapular nerve neurotization for optimal shoulder abduction and external rotation functional restoration. A total of seven ICNs were used to neurotize the axillary nerve in five children.
Gilbert39 reported that in his series, after ICN transfer to the axillary nerve, the results were disappointing. Samardzic40 stated that functional recovery (muscle grading M3 or more) was obtained in 63% of his posttraumatic brachial plexus palsy patients after axillary nerve neurotization with ICNs. In our series, three patients achieved an excellent result (muscle grading M4 to M5−) and two of them a good one (muscle grading M3+ to M4−) when axillary nerve was neurotized from ICNs, Fig. Fig.3.3. The initial signs of deltoid muscle recovery with this technique appeared between 6 and 8 months, and maximal recovery was achieved 14–24 months after the surgery.
Unlike adult posttraumatic brachial plexus injuries, hand reinnervation is of highest priority in the OBPP patient, as there is always the possibility of reinnervating the hand.2,24,41 Furthermore, hand sensibility is a precondition for future secondary reconstruction; therefore, restoration of the median nerve is important.41,42 Gilbert and Razaboni25 showed useful finger flexion (muscle grading M3 or more) in 75% of the cases, and useful intrinsic function in 50% after reinnervation of the lower trunk.
Kotani43 reported limited recovery of sensibility in 11 of 15 cases treated with ICN transfer for sensory restoration of the hand. Millesi36 showed recovery of protective sensation in 15 of 18 patients. Kawai30 reported superficial pain recovery and some light touch in 5 of 13 cases. Ihara44 stated that ICN neurotization of the median nerve provided some touch sensation in 12 of his 15 cases, but no two-point discrimination was recorded. Doi at al.45 showed sensory restoration after ICN transfer to the ulnar nerve in all patients. They found some recovery in two-point discrimination in 32% of the patients and return of tactile sensibility in 42% of them. In our series, ulnar nerve neurotization by ICNs was performed in seven children, and the median nerve was neurotized in three children. Nine out of 10 children obtained active wrist and finger flexion, and there was complete recovery of protective sensation in all children. In three patients in our series, we were able to record two-point discrimination on the ulnar fingers of the hand.
Rectus femoris muscle innervated from two ICNs was transferred as a free vascularized muscle to improve shoulder abduction function in one patient. Previously, during the first stage of brachial plexus reconstruction, the suprascapular nerve was directly neurotized by the spinal accessory nerve, and the axillary nerve was directly neurotized by T6 ICN, achieving fair results because of the long denervation time (36 months). After free rectus femoris transfer, the child achieved 65 degrees of shoulder abduction.
Akasaka and Hara46 reported that after free muscle transfer for biceps restoration, 8 of 11 patients achieved a muscle grading of M3 or more and had the ability for 80 degrees or more of elbow flexion.
In our series, gracilis muscle was transferred as a free vascularized innervated muscle in three patients to improve elbow flexion function. Approximately 6 months after gracilis muscle transfer for biceps restoration, muscle contraction corresponding to respiration was observed. One patient with a muscle grading of M2 preoperatively achieved a muscle grading of M4 postoperatively. Two other patients postoperatively achieved a muscle grading of M+3. All our patients exhibited more than 90 degrees of elbow flexion.
Three patients had free gracilis muscle transfer for finger extension restoration. All three patients reached a muscle grading of M+3 postoperatively and showed 10–20 degrees of finger extension.
Doi37 described their results after gracilis transfer for finger flexion restoration, with total active range of motion to be between 70 and 110 degrees in their patients. Two of our patients reached a muscle grading of M3, and four patients reached a muscle grading of M4 postoperatively, with the best muscle power found to be with the elbow joint in extension.
This boy was born at 40 weeks of gestation, vaginal delivery and vertex presentation. No forceps or suction were used. The baby's birth weight was 3900 g. During the preoperative examination, patient was unable to flex his right elbow. Patient was operated on at 26 months of age. Supra and infraclavicular brachial plexus exploration showed findings compatible with C5 rupture, C6 and C7 avulsion, and C8 and T1 rupture.
The T4, T5, T6, and T7 ICNs were used for direct neurotization of the musculocutaneous nerve. The suprascapular nerve was direct neurotize by the spinal accessory nerve. The proximal stump of C5 root was used to reconstruct the distal C8 and T1 roots.
Three years after the surgery, the patient has achieved full elbow flexion.
This girl was born at 40 weeks of gestation, vaginal delivery and vertex presentation. No forceps or suction were used to help the delivery. The baby's birth weight was 5200 g. During the preoperative examination, there was −80 degrees on right elbow extension. The patient was operated on at 32 months of age. Supra and infraclavicular brachial plexus exploration showed findings compatible with C5 rupture, C6 rupture, C7 avulsion, C8 rupture, and T1 traction.
The seventh, eighth, and ninth ICNs were used to neurotize the triceps branch. The suprascapular nerve was directly neurotized by the distal spinal accessory nerve. The posterior cord was reconstructed from the proximal stump of C5 root, and the C6 root was used for lateral cord reconstruction.
Two years after the surgery, patient has −20 degrees of elbow extension.
This boy was born after 40 weeks of gestation, with vaginal delivery and vertex presentation. The baby's birth weight was 4200 g. No suction or forceps were used to assist the delivery. The Apgar score was 5 during the first minute and 9 during the fifth minute. During his first office visit, the patient presented with the right upper extremity next to the trunk. He was operated on at the age of 10 weeks. The brachial plexus exploration revealed that C5, C6, and C7 roots were avulsed from the spinal cord, while the C7 and C8 roots were found to have suffered traction injury. The T6 and T7 ICNs were used for direct neurotization of the axillary nerve, and the T5 ICN was transferred to thoracodorsal nerve. The suprascapular nerve was reconstructed from the distal spinal accessory nerve through a 3-cm nerve graft.
Ten years after the nerve repairs, the patient has an excellent shoulder function, exhibiting a full abduction and external rotation.
ICN transfer is a worthwhile procedure for obstetrical brachial plexus reconstruction, especially in cases with multiple roots. Excellent results were seen in 85% of patients when ICNs were used for musculocutaneous nerve neurotization, and excellent results were achieved in 46% of patients after ICNs to triceps branch neurotization. Unlike the adult posttraumatic patients, after median and ulnar nerve neurotization with ICNs, almost all patients achieved protective sensation in the hand, and some of them achieved active wrist and finger flexion. Furthermore, free muscle transfers for elbow or hand animation can successfully be neurotized by ipsilateral ICNs to further enhance function in the paretic upper limb.