26 Congenital hand II
Disorders of formation (transverse and longitudinal arrest)
Repeated observation of the child is the most useful evaluation of the functional capability of each upper limb, aided by the Great Ormond Street Ladder (see Fig. 26.1). This may demonstrate capabilities not noticed earlier due to development, achievement of milestones and progressive myelination of the peripheral nervous system.
Most of these conditions are static with changes largely related to growth or surgical intervention. Early surgical intervention is indicated where deformity is progressive, for primary skeletal realignment or where growth will produce skeletal malalignment.
Key points 1
The birth of a disabled child is devastating for the parents. They will have many questions including those about the likely etiology of the condition. They may be looking for someone to blame. They frequently come with false hopes, encouraged by the media, about the possibilities of hand transplantation and stem cell technology. At this stage, the family may not be able to see beyond the child’s structural absence, to accept them as an individual in their own right.
The plastic surgeon may be able to assist in answering questions but, most importantly, they may act as a useful link to a disablement services team. This team of physiotherapist, occupational therapist, psychologist, prosthetist, orthotist and rehabilitation physician will be essential for the ongoing care of this child. They will help the parents accept their child and then, can look at ways of providing aids for the future. They may be able to put them in touch with other parents with children with similar disabilities or suggest parent run organizations which are specifically for the child with limb anomalies.
Occasionally, after full assessment by the rehabilitation team (Fig. 26.1), the plastic surgeon is asked to remove tissue/nubbins or make some surgical adjustment to the stump to make prosthesis fitting easier.
Sometimes, the parents themselves request to have the apparently useless tissue on the terminal part of the limb removed. When left alone they become favored by the child and there is a later reluctance to have them removed.
Traditionally this condition has been termed “amelia” (although strictly speaking from the Greek melos meaning limb this means the complete absence of the limb), peromelia, transverse hemimelia, or congenital amputation.
The level of transverse arrest is defined by the level of skeletal absence but there may be residual soft tissue beyond that point such as small nubbins. There are probably two main groups; those where there is a defect in limb formation and those where there is intrauterine amputation after limb formation.
Longitudinal development of a limb is dependent on the apical ectodermal ridge (AER). For the limb to fail to form, there has been an interruption in signalling with fibroblast growth factors (FGF 2, 4, or 8) to the progress zone which supports the AER, sometime in the period between week 5 and week 7 in utero, depending on the level of the transverse arrest.
With the exception of the teratogenic effect of thalidomide, etiological factors causing a failure of limb formation are largely unknown but may include genetic factors (there has been shown to be an autosomal recessive gene in Angora goats, which produces a deficiency of the distal limb segment1), chromosomal aberrations or environmental factors.
The patient presents with a congenital amputation (Fig. 26.2) which can occur at any level (humeral, proximal forearm, carpal or metacarpal). It may be bilateral – the left side is affected more at least twice as often as the right. Males are more commonly affected than females.
The diagnosis is usually straightforward at proximal levels but at the level of the carpus and metacarpals, there is frequently disagreement in terminology between transverse arrest and symbrachydactyly. Where there are distal nail elements or nail ghosts present on very short digital sacs, this represents symbrachydactyly but in the absence of these, the two conditions may be indistinguishable. If amniotic bands are present elsewhere, then the amputation is likely to be the result of a severely constricting band in utero. Rarely, a proximally situated amniotic band will have caused sufficient interruption to molecular signalling to create a distal hypoplasia which resembles a symbrachydactyly, with distal nail elements present. The management depends on the type of symbrachydactyly (Fig. 26.3).
Fig. 26.3 (A) Short finger type symbrachydactyly; (B) cleft hand type symbrachydactyly; (C) monodactylous type symbrachydactyly with free toe transfer; (D) peromelic type symbrachydactyly; (E) radiograph of peromelic symbrachydactyly.
With the advent of fetoscopic laser ablation for twin to twin transfusion, an iatrogenic cause of intrauterine limb loss should be considered where there is a history of this procedure. The mechanism is probably via disruption of the amnion creating bands rather than by the laser itself.2
Options for treatment here are between different prostheses, starting with a passive prosthesis. As the child develops active function, the passive prosthesis is replaced with a body-powered cable-controlled or myoelectric prosthesis.
Any treatment is usually restricted to functional and cosmetic prosthetics. The Krukenberg procedure3 has been used in this condition, particularly when bilateral, to provide a pincer grip between the two forearm bones. The interosseous membrane is extensively released and the defect skin grafted. This procedure has not been readily accepted by either surgeons or families because of the cosmetic deformity it creates in what is already an abnormal limb and is more appropriate in traumatic loss in adults where prostheses are unavailable and adaptation is more difficult.
Once again, prosthetics are the mainstay of management. From a surgical point of view, where there is sufficient carpus to be able to provide some intercarpal movement, a double free toe transfer may provide the ability to grip broader objects but is unlikely to be able provide a fine pincer movement due to the limited range of motion within the transfer. Inability to cup the palm, in the presence of a restricted motion distally in the toe transfers may limit the ability to provide a fine pinch.
Where all metacarpals are present, the simplest option is to deepen the first web to create some form of crude pinch (phalangization). However, although a simple procedure, the pinch is a crude lateral pinch, opposition is not possible and grasp is limited due to the poor span. There are two main alternative surgical options that may improve function further, each of which has a role and the decision to perform one rather than the other is frequently related to parental factors:
Free phalangeal transfer may be done at any age but it is claimed, that to obtain near normal growth within the phalanx, this must be undertaken within the first 2 years of life. However, there is no substantial evidence to support this. In our experience, growth plates remain open even if transferred at up to 7 years of age. This procedure is only appropriate where there is a sufficiently large soft tissue envelope and does best where the metacarpals form a normal cascade rather than where the central metacarpals are more deficient, forming a V shape.
First, a pocket must be created by dividing the longitudinal fibrous bands that extend to the tip of the soft tissue nubbin from the flexor-extensor tendon confluence over the metacarpal head. It is only then that the true extent of the soft tissue envelope becomes apparent. The surgeon needs to be wary of those nubbins with a constricted base that will not provide sufficient width to accommodate a phalanx.
It is necessary to harvest a whole phalanx since a partial phalanx inserted into the pocket will be prone to resorption (Fig. 26.4A). The phalanx needs to be taken together with an intact periosteal envelope but the authors do not try to preserve the volar plate and collateral ligaments (Fig. 26.4B). The phalanx is inserted into the soft tissue envelope (Fig. 26.4C). The phalanx is sutured onto the flexor-extensor hood which covers the metacarpal head. The flexor/extensors are not divided. A single K-wire is passed through the phalanx and the centre of the metacarpal head. Four weeks later, this is removed and active and passive movement is encouraged in the new metacarpophalangeal (MCP) joints (Fig. 26.4D–F).
Fig. 26.4 (A,B) Free phalangeal transfer. Bone was harvested with periosteum intact; (C) bone was inserted into soft tissue pocket; (D,E) patients hand with symbrachydactyly post free phalangeal transfer; (F) range of passive motion post free phalangeal transfer.
The extensor and flexor tendons in the donor toe should not be sutured together but there is no consensus on how to deal with the toe donor site defect. It has been suggested that inserting an iliac crest bone graft to the toe at the time of harvest may limit the donor site defect in the longer term.
Distraction augmentation manoplasty is considered as a secondary procedure when the child is old enough to participate in the decision and comply with all that this involves. In our experience, 8 years old is the earliest age that this is likely. Not all patients undergo this stage and so results are in a highly selected group. For this to be successful, the length of the bone being distracted must be a minimum of 10 mm long, it must be stable on the metacarpal and, for the fingers, the range of motion in the MCP joints must be >60°. In the thumb, the carpometacarpal joint must be stable. A distraction frame is applied to the bone with two Kirschner (K) wires or threaded pins fixing each bone proximally and distally (Fig. 26.5A,B). A subperiosteal corticotomy is performed between the proximal and distal fixation points (Fig. 26.5C) and the bones are distracted by 4 mm. Postoperatively, distraction is begun after 1 week at 1 mm/day, with weekly radiographs to monitor progress (Fig. 26.6).
Once the required soft tissue length is obtained, a further procedure is performed to bone graft the resultant defect with bone graft from the metatarsals, harvested subperiosteally (Fig. 26.7). This rapid soft tissue distraction differs from distraction osteogenesis (callotasis) in the rapidity of distraction and need for bone grafting into the sheath of osteoid that forms. Its advantage is the shorter time needed with the distractor in place in bone on which it may have a tenuous hold. Other authors4 have suggested that in children bone grafting is unnecessary but that has not been our experience where awaiting bone formation has increased the number of complications with failure of fixation and resulted in some loss of the initial length obtained.
Free toe transfer is suitable in those cases where there are toes present that are suitable to harvest, the family will accept the loss of one or both second toes and there are no contraindications to microvascular tissue transfer. One or two toes may be transferred simultaneously in a single procedure or sequentially in two procedures. There is no known benefit in performing this surgery before the age of 2 years and most surgeons would prefer to do it later than this. The site at which the toes should be placed depends on the individual hand anomaly.
For the procedure, the hand must first be dissected out to find suitable tendons, nerves and vessels and decide on the length of these that is required from the donor. To approach the blood supply to the second toe it is easier to explore the first web, rather than trace the dorsal metatarsal artery on the dorsum of the foot. Here, it can be clearly seen if the toe has a dorsal or plantar dominant blood supply and the vessel can be dissected retrogradely, once identified, and its dominance established. Removal of the whole of the second metatarsal gives better access when dissecting the toe than leaving the base and also allows easy closure of the resultant cleft in the foot to leave an acceptable donor site. Ideally, there are two teams present so that one can close the foot, while the other proceeds with the transfer(s).
The toe is K-wired in place on the hand, tendon and nerve repairs are performed and microsurgical anastomosis of arteries and veins is carried out; the level of this and hence the size of the vessel is dependent on multiple factors but it is preferable to avoid the need for vein grafts but also to avoid too long a vessel that may kink and adversely affect flow.
After microsurgical anastomosis in the hand, the surgeon needs to take care to avoid tension in closure of the wounds and it is better to place a skin graft than accept tension that may compromise perfusion.
In terms of growth outcomes, it is often stated that the transfers should be undertaken below 18 months of age but in our series of over 100 free nonvascularized phalangeal transfers this did not prove to be the case and we found open epiphyses were maintained even in transfers up to the age of 7 years.5 It is impossible to comment on whether or not the growth plates close prematurely because in all the published series the numbers are simply not enough to make a reasonable assumption. The viability of non-vascularized phalangeal transfers is assured provided they have a periosteal covering and there is no tension in the digital sac.
Where the metacarpals are of equal length with a good soft tissue envelope, free phalangeal transfers remain quite stable and can achieve up to 90° of motion at the new “MCP joint.” However, with uneven lengths of metacarpals, the direct placement of free phalangeal transfers onto the metacarpal heads will often lead to angulatory deformity and subluxation.
The secondary phase of distraction augmentation manoplasty is difficult and is prone to complications with regard to pin tract infections. The rapid distraction is relatively straightforward provided that the proximal phalanges are large enough to accommodate two pins proximal to the osteotomy site and two pins distal to it. Trying to attempt to distract multiple digits simultaneously using the same pins and a single fixator tends to compromise one digit whereas multiple fixators may not be feasible because of their bulk.
With regard to the digits that are produced this are often thin and more prone to fracture, but the children concerned value the digits enormously (Fig. 26.8). From a functional point of view, they improve the control of the flat hand, span and grasp of large objects. Psychological benefit to these children is from having four digits of equal length which move well at the MP joints. It is important to understand that the MCP joints that are created by the nonvascularized phalangeal transfers do achieve up to 80° of motion frequently.
Long-term follow-up is essential and it has become apparent to us recently that although the donor site in the toes initially shows surprisingly little deformity and looks satisfactory for up to 7 years, after 7 years very significant deformities can make themselves apparent. This is especially in terms of shortening, which only deteriorates in adolescence and functional issues related to shoe fitting may occur. These may result in toe amputation.
Bourke and Kay6 utilize a dowel-shaped bone graft from the iliac crest, with its apophysis, to replace the proximal phalanx in the foot and this may well be a very sensible approach but requires long-term study to assess its efficacy.
Three-quarters of the children will be expected to undergo secondary procedures including tenolysis and pulp debulking.7 Despite tenolysis, the active range of motion remains significantly less than the passive range although it appears better when the transfer is done in the older child.8 Recovery of protective sensation in the transferred digit is expected although, two-point discrimination and light touch sensation appear to recover better when the transfer is under the age of 8 years. The transfers are naturally incorporated into use, although their grip and pinch strength are less than on the normal side.9
The majority of children, when there has been appropriate counselling and discussion prior to surgery, have physical and psychological benefits from toe transfer (Fig. 26.9). Problems in the child may be anticipated where the parents are poorly adjusted to the hand anomaly.10
The donor site morbidity from free toe transfer depends on which toe is chosen and the number of toes taken from each foot. The authors do not harvest more than one toe from each foot where the foot is to be preserved. Their preference is the second toe transfer because, after closure of the defect as a ray amputation, this provides an excellent aesthetic and functional foot, particularly if the harvest is bilateral. Gait does not appear to be affected in the long term.
Occasionally, the prosthetist or the child will request removal of nubbins to make prosthesis fitting easier or for cosmetic reasons – this is a remarkably infrequent occurrence as prosthetic fitting has improved hugely and the nubbins are sensate and often used by the patient. They may request removal due to cold intolerance.
DAM may be requested after previous free phalangeal transfer to increase the length of the digits to increase span and help with specific tasks. This may result in the web spaces being expanded distally and so further web release may be required once distraction is completed. If a child is likely to want DAM, it is best to defer web release until after this is complete.
Tenolysis and pulp-plasty may be required in the majority of free toe transfers to improve the range of motion of the digits allowing fine opposition and to make the appearance of the toes on the hand less toe-like. The overall range of motion remains limited (average 40°) and is usually less than can be achieved with FPT but it is within a useful functional range.
Key points 2
Key points 3
Phocomelia describes an intercalary failure of formation, where there are distal structures present but a long bone, such as the humerus or radius and ulna, is largely absent (Fig. 26.9). Short finger type symbrachydactyly represents an intercalary deficiency but with smaller bones absent – these cases are not included usually within the term phocomelia.
The term phocomelia is derived from the Greek for seal limb. Although a proportion of the limbs affected may have a flipper-like appearance, many do not and others do not fit into this category – the term is therefore a poor one.
Phocomelia hit the headlines in the late 1960s when a rise in the number of children born with this and other congenital limb anomalies was noted and associated with the ingestion of thalidomide. The drug was subsequently banned from use and has only recently regained its popularity as an antiemetic, now used in those undergoing chemotherapy for malignancy. The children of those affected by thalidomide do not themselves appear to be at increased risk of producing children with limb anomalies.11
Although phocomelia is mainly sporadic, there are some problems with the same underlying genetic basis which produce four limb abnormalities which may be phocomelic. This includes Roberts syndrome, which is caused by mutation in the ESCO2 gene on chromosome 8, inherited in an autosomal recessive fashion.12 The ESCO2 genes are important in producing the ESCO protein product required for attachment of sister chromatid cohesion during S phases to allow chromosomal separation during cell divisions. In Roberts syndrome, the protein is abnormal so the chromatids attach poorly and cell division is delayed.
Phocomelia has been mimicked in chick limb buds by exposure to X-irradiation. It was thought that this caused a patterning defect as the cell’s identity was determined by its time spent in the progress zone. More recently, research has suggested that both thalidomide and X-irradiation cause defects due to a time-dependent loss of skeletal progenitors which do not survive or differentiate.13 It is known that antiangiogenic analogues of thalidomide induce chick limb defects. It is likely that thalidomide prevents angiogenesis and this would be expected to cause upstream changes in limb morphogenesis.14
Both upper limbs are affected, although this may be to different degrees. When all four limbs are affected, the prognosis is more guarded. Phocomelia may be isolated or may be associated with other more serious congenital anomalies that may be the deciding factor in overall prognosis.
The clavicle and scapula may be abnormal in addition to the glenoid, which is hypoplastic. In type III, it may be difficult to distinguish between a severe radial or ulna longitudinal deficiency and phocomelia especially in cases with TAR syndrome who typically have extremely short forearms. Many have abnormalities proximal or distal to the segmental defects which suggest that those cases considered to be phocomelia may, in fact, fit better into a diagnosis of longitudinal dysplasia (Fig. 26.10).15
In Roberts syndrome, there are severely shortened segments of all limbs especially the forearms and lower legs. The upper limbs are usually more affected than the lower limbs. There may be knee and elbow contractures. There is not always the full complement of digits, and those present may be abnormal. There are facial dysmorphisms, including microcephaly hypertelorism, downward slanting palpebral fissures, micrognathia, cleft lip and palate, a beaked nose, and small nostrils. There may be cerebral, heart, kidney and genital anomalies. Overall, growth is slow pre and post natally and there is some intellectual impairment. SC phocomelia used to be thought to be a separate entity but is now considered to be a milder variant of Roberts syndrome.
There is little place for surgical intervention here, except when excessive bone growth interferes with orthotic or prosthetic fitting. The occupational therapist and orthotist are key members of the team involved in finding ways to maximize the child’s independence and functional abilities. There may however, be some activities where they need assistance. This is more likely to be the case where the lower limbs are also affected.
Many children with normal lower limbs will use these to perform tasks for which the normal child would use their upper limbs. Many will use prosthetics or orthotics for certain specific tasks and discard them in favor of the remains of their sensate upper limbs and supplemented by their lower limbs for the rest of the time. Despite the restrictions of insensibility and weight that cause them to discard the functional prostheses, they may still desire non functional prosthetic arms for cosmetic reasons on occasion.
Any surgery carried out should be with the aim of allowing orthotic and prosthetic fitting and use as soon as possible afterwards. Primary wound healing without the need for skin grafting or complex bony external fixation is required.
Key points 4
• Radial ray dysplasia is a spectrum from minor size discrepancy in the thumb to complete absence of the thumb and radius. In any case of thumb hypoplasia, the proximal forearm and the contralateral thumb should be carefully evaluated.