Chapter 2
Nerve surgery for OBPL
A historic overview
31 Historic overview
I
n the early 1900’s, great interest developed in the treatment of peripheral nerve lesions, both adult traumatic brachial plexus injuries and birth-related nerve inju- ries. The first detailed repair of a stretch injury of the brachial plexus in adults was reported from Manchester in 1900.1 Three years later, the first nerve surgery for OBPL was reported by Kennedy from the University of Glasgow.2 Only three months later, on the other side of the Atlantic Ocean, a New York surgeon, Taylor, performed his first nerve surgical procedure for OBPL.3 He appears to have knowledge of the work of his Scottish colleague: Kennedy is cited in Taylor’s paper.Surgery for OBPL was apparently quite common at that time. For instance, neuro- surgeon William Sharpe (a trainee of Cushing) reported that 146 infants were oper- ated between 1913 and 1923 in the New York Polyclinic Hospital.4
The main focus of the present chapter will be the rise and fall of nerve surgical treatment of OBPL at the beginning of the twentieth century. Of special interest is the question why nerve repair surgery was abandoned after the initial spate of enthu- siasm. In addition, a brief history of the first descriptions of OBPLK will be provided, and the research on the causation of the condition will be summarized.
Erb’s palsy as a clinical entity
Although the condition is frequently referred to as Erb-Duchenne paralysis or Erb’s palsy, it is probable that other authors have described the occurrence of an obstetri- cal brachial plexus palsy earlier than Duchenne (1855) or Erb (1874).3,5 There is a gen- eral consensus that the first description of an OBPL was provided by the obstetrician Smellie in 1752.3,5-7 A spontaneously resolving paralysis of both arms was reported that lasted for several days. The cause was assumed to be prolonged compression which had occurred during tedious labour due to face presentation.8
Eventually in 1855, Duchenne described four cases as a distinct clinical entity. He depicts the typical clinical picture of a flaccid palsy in which abduction and external rotation of the arm were paralyzed, together with absent elbow flexion and supina- tion of the forearm.9 Twenty years later, Erb described a similar paralysis in adults.10 Using electrical stimulation in healthy subjects, he found a point on the skin of the supraclavicular region, about two fingerbreadths above the clavicle and about one fingerbreadth outside the sterno-cleido-mastoid muscle, which evoked a response of all the muscles found to be affected in this specific form of paralysis.10 This point cor- responds to the position of the fusion of spinal nerves C5 and C6 to form the superior trunk. It was concluded, therefore, that the causative lesion in such a paralysis was probably localized at this point.
Cause of the lesion
In his original publication, Duchenne attributed the palsy to pressure on the nerves by the fingers of the obstetrician or alternatively by the forceps during delivery. An alternative explanation was compression of the brachial plexus between the clavicle and the first rib.
To gain insight into the cause of the lesion, biomechanical experiments were per- formed with stillborn fetuses or infants who died shortly after birth. The New York surgeon, Taylor, performed such experiments during twenty dissections of the bra- chial plexus in ten children and he found that “only one thing caused stretching of the nerves, namely, increase in the distance from the neck and head to the shoulder”. He concluded that it was evident that such traction was the cause of the lesion. In addition it was described that the fifth root ruptured first, then the sixth, “and so on down the plexus in regular order if the force used was sufficient”. The resulting nerve lesion stretched over a length of sev- eral centimeters.3
Sever also claims to have performed “numerous dissections on infantile cadavers” in which he showed that traction and forcible separation of the head and shoulder puts the upper cords, the fifth and sixth cervical roots of the brachial plexus, “under danger- ous tension (…) like violin strings”. Furthermore, he described that forcible abduction and elevation of the arm and shoulder stretched the lower cords of the plexus, something which could happen in cases of breech delivery, with arms extended.5
In the Dutch medical literature, experiments on 17 still-born babies were reported in 1927. The results support the earlier notion that traction to the brachial plexus re- sulted in the typical lesion pattern of the upper roots of the brachial plexus.11
Histopathological samples from surgical cases were described. Torn nerve fibres and haemorrhage in the perineurial sheath of the upper cords of the brachial plexus were observed.3 This was later confirmed by other authors.12
Apart from these stretch injuries of the post-ganglionic nerve and the resulting neuroma formation, avulsion of the nerve root from the spinal cord was recognized as a specific pathophysiological mechanism. A case was described of bilateral palsy following a difficult breech delivery. The child died on the fifth day due to a cerebral haemorrhage and during autopsy spinal nerves C5 and C6 were found outside the dura.13
All these studies conclude that traction to nerves, leading to a stretch injury, could explain the observed clinical picture. The century-old description of the traction in- jury can be considered to be accurate. In the late 1970’s, biomechanical studies were again performed, the conclusions being similar to those described in the historical pa- pers.14 Currently, the pattern of injury is well recognized by nerve surgeons who treat such brachial plexus lesions in infants as well as in adults.
The emergence of nerve surgical treatment
In February 1903, Kennedy reported the first description of nerve surgery for an OBPL (the surgery had been performed in 1902).2 Two of his three cases were surgi- cally treated in early infancy (at 2 and 6 months of age), one case underwent delayed surgery at the age of 14 years. In this particular case, the main reason for surgery was torticollis, but the brachial plexus was also repaired during the operation. One year later, Kennedy had extended his surgical series to 5 OBPL cases.15
33 Historic overview
The surgical procedure is described in good detail and the surgical findings were reported as cicatricial changes of the superior trunk. The lesioned upper trunk was resected and the three distal divisions (suprascapular nerve, anterior division and posterior division) were sutured with a central suture of “fine chromotized catgut” to the fifth and sixth nerve. Because the resection of the upper trunk resulted in a gap, the shoulder of the infant was pushed upward and the head tilted to the side being oper- ated in order to be able to suture the stumps together. (Figure 1)
Kennedy claimed that 4-8 months after surgery the infants had fairly good use of their arm. Even the patient operated at the age of fourteen showed improvement, although this was not apparent until two years after surgery.15 (Figure 2)
Taylor described seven surgical cases in 1905.3 The surgical technique applied is ap- proximately the same as Kennedy’s, although the actual nerve suture used “lateral su- tures of fine silk involving the nerve sheaths only”.
Despite short operation times due to absence of proper anaesthesia, and the lack of proper illumination and magnification, both these surgeons were able to identify the same findings which are observed today.
Figure 1: Post-operative splint13
Note the abduction of the arm and lateral flexion of the head to the operated side which is necessary for a direct nerve suture
In Taylor’s patient group, mainly older children (4½ to 11 years) were operated: the three youngest were 8, 16 and 25 months. Of these three younger children, two died within days after the operation, resulting in a mortality rate of almost 30% in his total group. In a 1921 paper, Taylor describes having operated on 76 cases of obstetric palsy in total. In his last 25 patients (surgically treated from 1914-1921), only one patient
“died of hemorrhage on table” and in one case the surgery was stopped because of haem- orrhage.13
One of the difficulties was “to determine (…) whether the lesion will be likely to be sponta- neously recovered from or not. In some cases recovery takes place completely, although at birth all the typical signs are exhibited, and it is therefore necessary to wait a reasonable time in all cases before recommending operation.”15 In those days, electrical muscle testing was already employed to determine prognosis: “If (…) after two months no responses can be got in the mus- cles with the faradic current, (…) it is safer to proceed with the operation than to put off further
Figure 2: Historic photographs of results from Kennedy2,15
a b
c d
a) pre-operative; b) nine months post-operative; c and d) 2 years post-operative
35 Historic overview
time in the hope that recovery will eventually be the result.”2 This view was shared by Sher- ren in his 310-page monograph on nerve lesions: “Spontaneous recovery has taken place in about 70 per cent of the cases that have come under my observation. In many the paralysis had completely disappeared by the time the child was brought to have its electrical reactions tested at the age of three months. Complete spontaneous recovery rarely takes place if no improvement is noticed by this date.”16
Both Kennedy and Clark employed direct suture of the nerve ends after resection of the neuromatous tissue. The nerve ends were placed next to each other by means of external manipulation.
It is still a matter of debate whether a neurorrhaphy under such tension indeed resulted in good recovery. Today’s golden standard – using autologous nerve grafts to bridge the defect – was not commonly used in those days, although judging from publications the technique did seem to be known.17 Apparently nerve transplantation had already been applied in 1876 using a piece of nerve from an amputated limb. Also a dog’s sciatic nerve had been sutured in place of a nerve defect. The results of nerve transplantation were poor, and so nerve grafting was not widely applied in clinical practice.
Other reported surgical techniques include nerve transfer (which was then alter- natively named nerve crossing), fascicular transfer and end-to-side repair.17 All sur- gical techniques which one might consider to be new and modern, were apparently already known one century ago.
The first report, for instance, of end-to-side nerve repair of the brachial plexus in adults appeared in 190318, describing that surgical repair might “be effected by dividing the paralysed fifth cervical root and suturing its distal end into a neighbouring healthy root, the sixth or seventh.” To what extent this end-to-side repair yielded any clinical result was not reported. Kennedy was apparently aware of this end-to-side technique but did not use it in OBPL infants: “I should hesitate to implant the ends into one of the neighbouring nerves in this region, on account of the importance of the only nerves into which the implantation could be made with prospects of success; for, in the event of failure, the result might be that addi- tional muscles would participate in the paralysis.”2
Also, transfer of the accessory nerve to the brachial plexus was already described by Harvey Cushing. “In the upper part of the cord the fifth cervical segment was particularly the seat of injury, the deltoids being absolutely, and the biceps and supinator muscles in large meas- ure paralyzed. The spinal accessory has been sacrificed and transplanted into the upper radicle of the brachial plexus.”19
The main shortcoming of all surgical papers dating from the beginning of the twentieth century is the lack of substantial patient series that were accurately de- scribed postoperatively. There was no summary of cases by one or several authors, and so it is not clear exactly what the results of these surgical treatments were.
Nerve surgical treatment of OBPL declines
In 1916, orthopaedic surgeon Sever discards nerve surgery in a paper, in which he pre- sents 471 conservatively treated patients.20 On the basis of this personal series (which had grown to 617 in 1920 and to 1100 in 1925), he concludes “In regard to the operation on the plexus in the usual upper arm type of case, it might be said that in the experience of this clinic it has not been found necessary. (…) it cannot be too strongly emphasized that no operation on the plexus will be of any great use in restoring functional activity to the arm.”5
Sever was also pessimistic about nerve surgical treatment of total lesions, usually consisting of the avulsion of multiple lower roots from the spinal cord. He recognized that “the outlook is not so good, although many of the patients regain use of the upper arm in spite of the persistent paralysis of the lower arm and hand.” Despite the poor prognosis of sponta- neous recovery, nerve surgery did not improve the outlook: “it has been done a number of times without any benefit. The plexus in all cases was found to be so badly torn and so bound down and invaded by scar tissue that any kind of repair was impossible.”
Other authors shared the pessimistic results of nerve repair. “There has been no case yet (…) which has shown an anatomic and physiologic cure from the plexus operation. Even marked improvement is usually lacking. (…) Many times the nerve is so badly damaged that it is beyond repair.”21 Also neurosurgeon Sharpe remarks that “there is not one case of complete recovery of function” in his series of 146 patients.4
These conclusions are, however, not illustrated by descriptions of the neurological result in patient series which makes it difficult to judge the true results of the nerve surgery, but these were, most probably, poor. It is, therefore, understandable that the voices of the initial enthusiasts simply became weaker and the voices of the oppo- nents of nerve surgery became stronger.
Instead of repair of the nerve lesion, Sever and Jepson advised treating the seque- lae of OBPL patients by orthopaedic surgery. Sever recommended performing muscle and tendon transfers for improvement of shoulder function, together with release of restricted shoulder mobility.20 Jepson favoured a rotation osteotomy to restore func- tional use of the limb.21 The results of these surgical strategies were claimed to be su- perior to nerve repair. Modifications of these secondary surgeries are still performed today.22
Another factor which may have led to the decline of nerve surgery was the associ- ated mortality. In those days, ether was used to produce anaesthesia. In the early se- ries of Taylor, a mortality rate of 2/7 (almost 30%) was reported.3 Other surgical series do not report such high mortality rates, but there must have been a certain degree of mortality resulting from surgery in such young infants.
Nerve surgery gradually lost popularity. “Taylor, who has had a considerable experience of operations directed to the damaged nerves, now (1938) prefers to wait as long as any improve- ment is taking place.”23 Finally, nerve surgery for obstetric lesions was abandoned for a considerable period.
37 Historic overview
Fifty years later, peripheral nerve surgery of lesions in adults was renewed by introduc- ing technical improvements: the operating microscope, better surgical techniques, improved suture material, and in particular the emergence of the use of autologous nerve grafts. Together with improved safety of anaesthesia, nerve surgery for infants was re-introduced in 1978 by Alain Gilbert.24
Summary
At the beginning of the twentieth century, nerve surgery for OBPL was performed for the first time, gaining popularity in the subsequent two decades. The origin of the lesion was investigated and discussed thoroughly. It was recognized that the major- ity of infants demonstrated spontaneous recovery, but that for a subgroup, surgery might be warranted. Additionally, in those patients with a total lesion of the brachial plexus, spontaneous recovery of hand function was known to be poor. Selection for surgery was based on clinical grounds, sometimes together with the results of electri- cal stimulation.
Despite limited illumination, very short operating times, and lack of proper anaes- thesia, these pioneer surgeons were able to deliver an accurate report of the pathology, which corresponds to current surgical findings.
Although some papers from this era systematically summarize cases and the re- sults of surgery, these series lack valid outcome measures to compare surgical therapy to other treatments or to the outcome of spontaneous recovery. The advocates, as well as the opponents of surgery, only report the numbers of patients in order to gain authority as experts; the detailed results of their treatment were not published.
Surgery was abandoned because its usefulness could not be demonstrated, be- cause of the reported mortality, and because alternatives, i.e. secondary surgery, be- came available.
1 Thorburn W. A Clinical Lecture on Second- ary Suture of the Brachial Plexus. Br Med J 1900 May 5;1(2053):1073-5.
2 Kennedy R. Suture of the brachial plexus in birth paralysis of the upper extremity.
Br Med J 1903 February 7;1(2197):298-301.
3 Clark LP, Taylor AS, Prout TP. A study on brachial birth palsy. Am J Med Sci 1905;130(4):670-707.
4 Sharpe W. End results in neuro-surgery:
impressions during decade 1913-123. Ann Surg 1925 November;82(5):684-97.
5 Sever JW. Obstetric paralysis. Its cause and treatment. Can Med Assoc J 1920 Febru- ary;10(2):141-61.
6 Harrenstein RJ. De geboorteverlamming van den arm [in Dutch]. Ned Tijdschr Geneeskd 1922;66:1435-47.
7 Ashhurst APC. Birth injuries of the shoul- der. Ann Surg 1918 January;67(1):25-50.
8 Smellie W. Smellie’s Treatise on the theory and practice of midwivery (edited with annotations by McClintock, A.H.). London:
The new Sydenham Society; 1876.
9 Duchenne de Bologne GB. De l’Électrisation Localisée et de son application et a la thérapeutique [in French]. 2 ed. Paris: J.B.
Baillière et fils; 1861.
10 Brody IA, Wilkins RH. Neurological Clas- sics: Erb’s palsy; “On a characteristic site of injury in the brachial plexus” by W. Erb 1874.
Arch Neurol 1969 October;21(4):442-4.
11 Harrenstein RJ. Proefnemingen en practische ervaringen over de geboorte- verlamming van de arm [in Dutch]. Ned Tijdschr Geneeskd 1927;828-46.
12 Barlow T, Ballance CA. Rupture of the Upper Cord of the Brachial Plexus at Birth. Proc R Soc Med 1908;1(Clin Sect):215-20.
13 Taylor AS. So-called congenital dislocation of the shoulder posterior subluxation. Ann Surg 1921 September;74(3):368-75.
14 Metaizeau JP, Gayet C, Plenat F.
Brachial plexus birth injuries. An experi- mental study. [in French]. Chir Pediatr 1979;20(3):159-63.
15 Kennedy R. Suture of the brachial plexus in birth paralysis of the upper extremity.
Br Med J 1904 October 22;1065-8.
16 Sherren J. Inuries of nerves and their treat- ment. London: James Nisbet & co.; 1908.
17 Sherren J. Some points in the surgery of the peripheral nerves. Edinburgh Medical Jour- nal 1906 October;20:297-332.
18 Harris W, Low V.W. On the importance of accurate muscular analysis in lesions of the brachial plexus and the treatment of Erb’s palsy and infantile paralysis of the upper extremity by cross-union ofthe nerve roots.
Br Med J 1903;2:1035-8.
19 Cushing H. The special field of neuro- logical surgery. Bull John Hopkins Hosp 1905;16(168):77-87.
20 Sever JW. Obstetric paralysis. An ortho- pedic problem. J Bone Joint Surg Am 1916;s2(14):456-75.
21 Jepson PN. Obstetrical paralysis. Ann Surg 1930 May;91(5):724-30.
22 Mehlman CT, Koepplinger ME. Hyphenat- ed history: the Sever-L’Episcopo procedure.
J Pediatr Orthop 2007 July;27(5):533-6.
23 Morison JE. Peripheral brachial paralysis in infants and Children. Arch Dis Child 1938 December;13(76):310-32.
24 Gilbert A, Khouri N, Carlioz H. Birth palsy of the brachial plexus -surgical exploration and attempted repair in twenty one cases.
[in French]. Rev Chir Orthop Reparatrice Appar Mot 1980 January;66(1):33-42.
