Late Presentation

Late Surgery, Nerve transfers, Neurotizations, Delayed late repair


The early repair of OBP ( Obstetrical Brachial Plexus) has been well standardized . Early surgery is the gold standard , although there are sometime differences in the timing. When we talk of early surgery , it encompasses the commonly used 3-month term for decision , up to 6 or 9 months for some surgeons. It is widely accepted that the possibilities are still good of recovery even after 9 months and up to 12 months and these different choices come from various reasons but not the potential degradation of the result due to a late repair.

There is a grey zone between 12 and 18 months where a repair seems still to be possible but where the potential recovery may not be as good. There are no scientific papers studying the situation.

After 18 months it is widely considered that the repair will not lead to an acceptable result , although there are situations where it may still be indicated ( late , after 3-5 years , repair of complete paralysis). Recently , the advent of neurotizations has probably changed the concept : repairs purely motor , close to the muscle , coming from a normal nerve . The fundamental question is : how late can the muscle be reinnervated and still give a reasonable function.


There are various situations where the patient is seen late and needs some kind of repair :
  • late presentation or incomplete recovery.

    The family had no exposure to a competent team or has refused previous propositions of repair . There is usually some kind of recovery , except in avulsion injury of the upper roots. The recovery is weak ( Biceps) or incomplete ( external rotation ) . There are possibilities of late repair in these cases . Delayed grafting ( over 18 months ) does not usually give very good results. The other drawback of the graft is the obligation to resect the neuroma , thus loosing the little function recovered. It is difficult to gain acceptance from the families. On the contrary , the neurotizations are easily accepted as they do not imply the loss of function. The most commonly used neurotizations in babies are the Spinal Accessory to Suprascapular nerve transfer and the branches of Ulnar and/or Median nerves to the Musculocutaneous branches. The Spinal Accessory transfer is straightforward but the arm transfers may be tricky . It is first necessary to explore the Musculocutaneous nerve and its branches. Each branch is dissected and stimulated with low voltage stimulation . In late presentations there is often a weak elbow flexion. It is important to determine whether the flexion is due to the Biceps or the Brachialis. If both muscles give no answer, they will be both neurotized. If one of the muscles seems better , it can be left and the other neurotized.
    Both Ulnar and Median nerves are stimulated . Although in upper lesions the Ulnar nerve should give a stronger answer than the Median nerve , the reality is sometimes different .The choice of donor nerve is based on the quality of the response and the proximity to the recipient nerves. The branches to Biceps and Brachialis can be dissected into the Musculocutaneous nerve for 3-5 cm giving a good length to the pedicle . If the patient has a good Triceps and a week Deltoid , a nerve transfer from one of the branches to the Triceps to the upper axillary nerve is always added. This is done using the same incision , in the axilla . This approach allows a good transfer without any tension .

  • Incomplete recovery after surgical repair.

    There are cases where the child is seen after a previous repair , usually with grafts, of the plexus. The usual time of recovery after a graft is around 8-12 months for the Biceps and 18 months for the external rotation. If nerve transfers were done, the time of recovery is much quicker ( 5 to 8 months). There is then a moment, after sometimes 2 years where the recovery has stopped and the possibility of a new nerve repair is discussed. The absence of fibrillations precludes in our opinion the possibility of a new repair . In fact , when there has been a repair , there is always some signs of regeneration in the affected muscles. The muscles seem to stay alive much longer .

  • Results after late repairs.

    Delayed grafting ( over 18 months ) does not usually give very good results. The other drawback of the graft is the obligation to resect the neuroma, thus loosing the little function recovered . It is difficult to gain acceptance from the families. On the contrary , the neurotizations are easily accepted as they do not imply the loss of function. Delayed transfers seem to give much better results. With P. Raimondi we presented at the Narakas Club in Lausanne ( 2013 ) a follow up of 53 patients operated for late neurotizations in adults and children. Of 24 late neurotizations ( more than 12 months) of the elbow flexors in children, we found :2 M0 , 4 M1-2 , 18 M3-4 (75%). Of 15 neurotizations of the Suprascapular nerve , we found 2 M0 , 6 M1-2 , and 7 M3-4 (47%).

Result after late repair with nerve transfers


There are again several clinical pictures . The most common is the complete paralysis with spontaneous recovery . This picture is very common in underdeveloped countries . The child is seen after 18-20 months with often a good biceps , an incomplete shoulder and a completely paralysed hand , often stiff , in supination . This terrible situation is extremely difficult to treat .There are no palliative procedures that can give a functional hand in a complete paralysis .
The electrical studies will determine the reconstructive possibilities . The presence of fibrillations in the flexor muscles allow to hope for a possible reinnervation . The difficulty being the quality of the donor nerve .

  1. There is a good proximal recovery in particular of the elbow flexors .

    Exploration of the plexus and grafting from the proximal upper roots is a possibility but in most cases it is impossible to obtain a consensus from the family . Loosing every recovered muscle with no certitude to recover more ( the hand and the elbow and shoulder) will not be accepted . It is necessary to bring the donor nerve closer to the target . In that case , the neurotization using the nerve of the Brachialis muscle is a good option : there is no loss of elbow flexion , and in many cases the suture can be done end to end , without graft .
    The incision starts in the arm where the branches of the Musculocutaneous nerve are found and stimulated . To be allowed to use the branches to Brachialis , one must be sure of the quality of the Biceps reinnervation . The branches to the Brachialis ( usually 2 ) are dissected to the muscle . If they are of good caliber , it is possible to use only one , preserving the function of the muscle .
    In the forearm , the Median nerve is dissected distally , with all its branches to the Flexor muscles .The priority is to reinnervate the Anterior interosseous nerve but the Flexor Carpi Radialis is also important as the Flexor Profundus . Stimulation is very important as sometimes there is a partial recovery and those muscles functioning should be preserved .
    The nerves to reinnervate are followed under microscope inside the Median nerve . This intraneural dissection progresses up into the nerve and often it is possible to isolate these branches up to the lower arm , at the level of the Brachialis nerve , allowing a direct suture .
    In more than 2/3rds of our cases , we have been able to do a direct suture . In the other cases , the Medial cutaneous nerve is used as a graft .
    This operation has given in our hands approximately 50% of useful results ((8/17). It is based on the capacity of a muscle with fibrillations to be able to recover .

  2. There is no or weak proximal recovery .

    The shoulder can be improved by local neurotizations ( Spinal Accessory) . For the elbow and the hand , I now feel there is no other solution than the use of Contralateral C7 . It has been used in the past with the technique described by Gu (1992) in adults . The use in Obstetrical Palsy has also been reported ( Chen 2007; Lin 2010,2011 ). The results seem to be better with a excellent results on the upper trunk but only 40% of good results on finger flexors .
    The modification introduced by Wang has changed the results : his prespinal route for the C7 root allows to shorten the distance with the other side and in many cases to avoid grafting . The results are thus quicker and better in quality . The dangers of this retro esophageal route are real but a good technique should avoid them. (Wang 2013, 2014)
    We use an anterior approach to the spine as in the Cloward operation . This allows , on each side to control the vessels and to channelize the space between the oesophagus and the vertebrae . The C7 root can be dissected longer in the adult and allows most of the time to retrieve it on the affected side . It is often shorter in the babies and will then need the adjunct of a nerve graft to coapt with the lower trunk .This retro esophageal nerve graft is technically difficult to pass and to maintain.
    With this technique Wang ( 2014 ) reported 103 adult patients with a result of 66% of good results ( M3-M4) in the fingers flexors . This result is obtained in an average of 14 months .
    Another point is the defect created by C7 resection ( at the level of its bifurcation ) . All the authors have seen very commonly sensory defect in the median nerve distribution in adults and rare muscle weakness . These defects disappear almost constantly after 3 months . In our experience , we have had small defects ( weakening of Triceps or extensors , sensory anomalies in the hand in adults ) but all had recovered by 3 months .
    It is important to make a complete repair of the plexus : The Contralateral C7 will also be used to reconstruct the Musculocutaneous nerve ; when reinnervating the lower trunk , the axons grow through the Medial Cutaneous nerve . In the arm , this nerve is cut and anastomosed to the MC nerve ( Wang 2014) .In the same time , the Accessory to Suprascapular neurotisation will allow recovery of External rotation an some abduction . It is also necessary to reconstruct the posterior cord . This can be done using the Phrenic nerve .
    Although in adults using the phrenic nerve has usually no consequences , it may be different in small children . There is a risk in paralyzing the diaphragm and some very important conditions must be respected :
    • The child should not be too young . A diaphragmatic paralysis has more consequences at 3 months than 20months
    • The contralateral phrenic should be respected . Even a temporary sideration ?? is dangerous
    • The intercostal nerves transfer should not be associated

  3. It is too late for nerve reconstruction in the lower trunk : the muscles are denervated .

    Although there are some studies showing that appropriate stimulation may reverse denervation and allow reinnervation after a long time ( Kern 2002, Salmons 2005) , it is not yet clinically proved and is still a research path .
    At the present time , the only viable possibility for reconstruction of the hand function in a complete lower roots paralysis will be tendon or muscle transfers .
    Pedicled tendon transfers : It is rare that the proximal muscles are in good shape in a traumatic plexus but this happens more often in obstetrical paralysis . It may then be possible to use an extended Latissimus transfer It can be used for flexion of the fingers or for extension , in association with another transfer for the flexors . Or a pedicled Biceps tendon transfer . The Biceps can be transferred to the flexors but this technique has a few drawbacks : a transfer cannot work over 2 joints unless one of them is stabilized . It will not be possible to obtain a good fingers flexion if there is no Triceps ( or a splint in elbow extension ) .The other difficulty is the necessity of a pulley at the elbow . I use either the Brachioradialis or I try to maintain the Lacertus Fibrosus . • Vascularized muscle transfers .
    They are theoretically the ideal solution but the main difficulty is to find a good donor nerve .
    In the upper extremity various muscles have been used but the most adapted to the anatomy and needs is the Gracilis .The use of the Gracilis has been developed originaly for the treatment of Volkmann’s contractrure and applied to extensive paralysis . The main difficulty is the selection of a good donor nerve in this paralysed extremity . Some attemps have been done to use the contralateral Pectoral nerve in two stages ( Gilbert 1977 ) but the development of local neurotisations ( Spinal Accessory , Ipsilateral Pectoralis , Intercostals) has allowed a more convenient one stage operation . We have done 69 vascularized gracilis transfers from 1976 : 42 were done for upper limb paralysis ( 22 for Biceps , 1 for Triceps , 16 for finger flexion , 3 for extensors )
    The results are good for elbow flexion ( 66% M3+ ) but poor for fingers flexion and extension ( 24% M3+). This is mainly due to :
    • It is more difficult to find a distal nerve . Using a proximal nerve leads to a muscle crossing two joints .
    • To use properly the flexor muscle , an active extension is needed .
    • There is often stiffness in the fingers joints .