Evidence based Medical (& Surgical) Practice
Clinicians all over the world are becoming increasingly aware of the need for a more rational approach to medical treatment. Many new treatment modalities that have been reported in the literature are not based on sound scientific evidence.
Sackett1 suggested that the scientific "weight" of different types of studies performed for deciding treatment options are in the following order of descending value :
- 1. Randomised controlled trials
- 2. Non-randomised cohort comparisons of contemporaneous patients
- 3. Non-randomised historical cohort comparisons
- 4. Case series without controls.
While assessing therapeutic effectiveness of a surgical procedure the preferred study design is a randomised controlled trial (RCT). The Cochrane Collaboration has been preparing a database of all published RCTs. The Musculo-Skeletal Section of the Cochrane Collaboration had requested the Editor to scrutinise the literature published in India and identify all RCTs. Not a single study which fulfilled the criteria of a RCT could be located in Orthopaedic literature in India !
It is certainly incumbent on us to improve the quality of surgical care in the field of Paediatric Orthopaedics in India. This will only be possible if we make a concerted effort to base our treatment protocols on scientific evidence. The article of Harrison & Bassett abstracted in this issue clearly demonstrates how a RCT failed to show any therapeutic value of a treatment method which, initially, was thought to be beneficial.
It is our hope that POSI would lead the way among our Orthopaedic colleagues in improving the quality of clinical research with the aim of practising evidence based medicine.
1. Sackett DL. Rules of evidence and clinical recommendations on the use of antithrombotic agents. Chest 1986; 89: 2-35.
1. Management of fibular hemimelia - Amputation or limb lengthening.
Naudie D, Hamdy RC, Fassier F, Morin B, Duhaime M. Shriners Hospital for Crippled Children, Montreal, Canada. Journal of Bone & Joint Surgery (Br), 1997;79-B:58-65.
The study compares the results of 12 patients (13 limbs) who underwent early amputation and prosthetic fitting with those of 10 patients who were treated by lengthening using the Ilizarov apparatus. The twelve patients treated by amputation were functioning well with few complications, while the ten patients who underwent lengthening had to wear braces, shoes-raises or undergo further corrective surgery. Two of the ten children in the lengthening group had to undergo late amputation for poor function or cosmesis. There were fewer hospital admissions, follow-up visits, and interruptions of schooling in the amputation group.
The authors conclude that amputation is a more effective method of management than limb lengthening in severe degrees of fibular hemimelia.
The paper brings to sharp focus an issue which was briefly discussed at the last Meeting of POSI. Is limb salvage really the best option for children with tibial and fibular hemimelia ? No one would question the laudable aim of preserving these limbs. However, we also need to carefully evaluate whether by preserving a limb, we are preserving an acceptable degree of function that is normally expected of the limb. If, after a series of expensive, technically demanding and often painful operations, the function of the limb does not equal the function afforded by a prosthesis, an honest reappraisal of our approach is needed. The very fact that two patients in the series opted for an amputation later, should serve as a warning that mere equalisation of limb length may not be enough for patients with these anomalies.
The Editor is aware that there may be very divergent views on this issue. Readers are invited to send in their comments, based on their personal experience or on the literature, to the Editor, and this could be the basis of a debate in one of the forthcoming issues of POSITIVE.
2. Hemiepiphyseal stapling for knee deformities in children younger than 10 years ; A preliminary report.
Mielke CH, Stevens PM. University of Utah, Salt Lake City, Utah, USA. Journal of Pediatric Orthopaedics 1996;16:423-9.
The authors report the results of stapling one side of the epiphysis to correct genu valgum and varum in children under the age of 10 years (mean age: 6 years & 4 months). After a follow-up of three years they noted successful correction of the deformities without any evidence of permanent growth plate arrest. They conclude that this procedure is safe and effective for correcting angular deformites of the knee in children under 10 years of age.Editorial Comment
Despite the fact that stapling of the epiphysis is considered to be a temporary method of arresting epiphyseal growth, many surgeons reserve this procedure for children over the age of 8 years (as advocated by Blount). This is because of the fear that permanent physeal arrest may occur. This appears to be the first report which claims that it is safe to perform this procedure in young children. If the safety and efficacy of the procedure is confirmed in subsequent studies, it would be a simple and cheap alternative to corrective osteotomy, or the Ilizarov technique for pure angular knee deformities in young children.
3. Surgical management of ankle valgus in children: Use of a transphyseal medial malleolar screw.
Davids JR, Valadie AL, Ferguson RL, Bray EW, Allen BL. Shriners Hospital for Children, Greenville, South Carolina, USA. Journal of Pediatric Orthopaedics 1997;17:3-8.
Valgus deformity of the ankle secondary to a wide variety of conditions in 17 children (29 extremities) was treated by placing a single screw from the tip of the medial malleolus across the distal tibial physis. The tibiotalar axis was considered to be the best measurement to define the extent of ankle valgus. The median age at surgery was 11 years and two months. The median rate of deformity correction was 0.59 degrees per month. Resumption of physeal growth occurred after removal of the screw but recurrence of deformity occurred at the rate of 0.6 degrees per month if the screw was removed before skeletal maturity. The authors conclude that the technique is minimally invasive, minimally morbid, and technically a simple method of producing reversible partial physeal arrest at the ankle.
4. Screw epiphysiodesis for ankle valgus.
Stevens PM, Belle RM. University of Utah, Salt Lake City, Utah, USA. Journal of Pediatric Orthopaedics 1997;17:9-11
Progressive ankle valgus may develop insiduously in neuromuscular disease, skeletal dysplasia, chromosomal anomalies and clubfoot. This may occur concomitantly with or be mistaken for hindfoot valgus. In 31 children (50 feet) with ankle valgus deformities, a single 4.5 mm screw was inserted percutaneously from the medial malleolus across the physis as an outpatient procedure. Post-operatively, no immobilisation was required. Satisfactory correction of the deformity with low morbidity was achieved without permanent physeal closure. The authors conclude that this is a safe and predictable solution for progressive ankle valgus in children.Editorial Comment
Both these articles draw our attention to a deformity that may often be mistaken for hind foot valgus. A closer look at children with neuromuscular problems in our clinics would certainly show that ankle valgus is by no means a rare deformity. The papers also describe a very elegant method for dealing with this problem. The simplicity of the approach is what makes the technique most appealing.
5. Epiphysiodesis of the lower extremity : Results of the percutaneous technique.
Horton GA, Olney BW. University of Kansas Medical Center, Kansas City, Kansas, USA. Journal of Pediatric Orthopaedics 1996;16:180-2.
The results of 42 percutaneous epiphysiodeses of the lower extremity are reported. Physeal arrest was achieved in all the cases without any significant complication or development of angular deformity. The average duration of hospitalisation was 1 day. The technique entailed drilling of the physis with a 6mm drill bit under image intensification, through a very limited skin incision. Physeal ablation was completed with a curet under fluoroscopic control.
6. Epiphysiodesis using a cannulated tubesaw.
Macnicol MF, Gupta MS. Princess Margaret Rose Orthopaedic Hospital, Edinburgh, Scotland, UK. Journal of Bone & Joint Surgery (Br) 1997;79-B:307-9.
The authors describe a technique of epiphysiodesis using a cannulated saw. Under image intensification, the "saw", which is introduced by hand over a guide wire, cores out a plug of bone including a strip of the physis. The rest of the physis is curetted through the opening created by the saw. After radiographically confirming that the physis has been ablated, the plug of bone is reintroduced to hasten the epiphysiodesis and to reduce bleeding. The technique was found to be effective in achieving epiphyseal closure without the development of angular deformities at the knee in 35 children who were operated at a mean age of 13.5 years (range 10-15 years). At skeletal maturity, the mean predicted limb length discrepancy of 3.3 cm (range 2.0 to 4.5 cm), was reduced to 0.7 cm.Editorial Comment
Both these papers describe percutaneous techniques of epiphysiodesis. It is generally well accepted that percutaneous drilling of the growth plate is effective in producing physeal arrest and that it results in fewer complications than the open Phemister technique. The method described in the second paper seems an attractive modification of the percutaneous drilling method initially advocated by Bowen & Johnson (1984). The authors must be congratulated on being able to time the procedure accurately enough to achieve such a satisfactory degree of limb length equalisation. The success of epiphysiodesis depends on the correct timing of the procedure, which is based on calculations; which, in turn, are dependent on the accurate determination of the skeletal age. Skeletal growth atlases similar to those of Greulich & Pyle (1959) or Tanner et al (1975) which are applicable to Indian children are not available. Without such an atlas estimation of bone age would be inaccurate and this would make our results of epiphysiodesis unsatisfactory. It would be well worth the effort if POSI would take the initiative in generating data for such an atlas for Indian children.
7. Partial wound closure after surgical correction of equinovarus foot deformity.
Ferlic RJ, Breed AL, Mann DC, Cherney JJ. University of Wisconsin-Madison, Madison, Wisconsin, USA. Journal of Pediatric Orthopaedics 1997; 17:486-9.
The authors reviewed the effect on wound healing in 31 clubfeet in whom the medial skin incision was left open to heal by secondary intention. The wound was left open if it was felt that primary closure would compromise the circulation to the skin or if closing the incision would require loss of correction of the deformity. The skin edges were loosely approximated with a running "baseball suture". Cast changes were made under anaesthesia on the 7th and 14th day. All the wounds except one healed by six weeks. The appearance of the skin incision was similar to those in which primary wound closure was possible. There were no infections. The authors conclude that skin closure is not essential after surgical correction of clubfeet.Editorial Comment
The suggestion of leaving a wound open may seem quite unconventional to most surgeons. However, there seems to be merit in this approach which is certainly simpler than performing rotation flaps, Z-plasties, or using tissue expanders.
8. The results of a double-blind trial of pulsed electromagnetic frequency in the treatment of Perthes' disease.
Harrison MHM, Bassett CAL (deceased). Royal Orthopaedic Hospital, Birmingham, England. Journal of Pediatric Orthopaedics1997; 17:264-5.
21 boys with Perthes' disease were divided into two groups and treated by non-weight-bearing orthotic treatment. A pulsed electromagnetic frequency (PEMF) coil was applied over the hip. The coil was inactive in one group. The activity of the coil was unknown to the clinic staff and the patients. There was no difference between the two groups in the time taken for femoral head reconstitution.Editorial Comment
Though an earlier report by Harrison had suggested that PEMF may reduce the time taken for revascularisation in Perthes' disease, this prospective double-blind-trial disproved it. The study emphasises the need for more scientifically designed controlled trials such as this, to confirm or refute unsubstantiated impressions we surgeons have about the efficacy of treatment of various conditions. Impressions are not enough; we must have proof that the treatment we give is the best we can offer.
Management of Perthes Disease
This brief review will attempt to answer the following questions based on currently available literature and the author’s personal experience.Why treat Perthes’ disease ?
Perthes’ disease per se is a self limiting disorder characterised by avascular necrosis of unknown aetiology affecting the capital femoral epiphysis.Complete revascularisation of the avascular epiphysis occurs almost invariably, over a period of about two years, without any form of treatment whatsoever. In the light of this knowledge the need for any treatment for Perthes’ disease may be questioned.
However, during the process of revascularisation, in a significant proportion of patients, deformation of the epiphysis occurs if appropriate treatment is not given. In spite of this, the children become virtually asymptomatic in most instances. Once again, it may be argued that treatment is not needed as children with femoral head deformation at healing have no pain or functional disability. The follow-up of these children to adult life, however, clearly explains why treatment of Perthes’ disease during its active stage is necessary. Several young adults with healed untreated Perthes’ disease develop clinical and radiological features of secondary degenerative arthrosis1. It has been our experience that such patients present with pain by the early part of the fourth decade of life. Since treatment of osteoarthrosis of the hip in young adults poses a very serious problem, it is incumbent on us to attempt to avoid this late complication of Perthes’.The purpose of treating Perthes’ disease, therefore, is to attempt to prevent late secondary osteoarthrosis of the hip.
It is emphasised that the treatment of Perthes’ is not aimed at influencing the rate or completeness of revascularisation of the femoral epiphysis as the disease is self limiting and currently there are no known methods which appreciably hasten the revascularisation process.What is the rationale of treatment of Perthes’ ?
Since the sole purpose of treating Perthes’ disease is to prevent osteoarthrosis, we need to understand the factors which lead to this complication and attempt to prevent them. A closer look at the morphological changes in the hips of adults with secondary osteoarthrosis show that all of them had one or more of the following changes : an irregularly shaped femoral head (coxa irregularis), a large femoral head (coxa magna), or a short neck (coxa breva) with relative overgrowth of the greater trochanter. It is clear that joint incongruity seen in coxa irregularis would lead to osteoarthrosis of the hip. Both coxa magna and coxa breva result in altered mechanics of the hip leading to excessive stresses on the hip joint which in turn leads to osteoarthrosis.
It would be ideal if all these morphological changes could be prevented by treatment. In order to attempt to do so, we need to be aware of the pathogenesis of these structural changes in the proximal femur.Pathogenesis of coxa irregularis
During the first stage of Perthes’ disease, hypertrophy of the articular cartilage of both the femoral head and the acetabulum occurs as a consequence of the synovitis that follows the vascular insult. Maximal cartilage hypertrophy occurs medially, causing the femoral head to be displaced laterally. Consequently, the lateral part of the infarcted zone of the epiphysis extrudes beyond the confines of the acetabulum. During the stage of fragmentation, collapse of the necrotic trabeculae occurs. In the stage of regeneration and revascularisation, immature woven bone replaces the avascular bone. The new bone is vulnerable to deformation until it is converted to lamellar bone. At this stage, muscular forces and weight-bearing stress transmitted across the acetabular margin cause the “biologically plastic” woven bone in the extruded part of the femoral epiphysis to deform.Pathogenesis of coxa breva and coxa magna
The vascular insult, if severe enough, can lead to damage to the growth plate of the capital femoral epiphysis resulting in premature fusion. The consequence of this would be a shortened femoral neck or coxa breva. As the greater trochanteric apophyseal growth plate is not affected, the trochanter continues to grow and this results in what is termed “greater trochanteric overgrowth”. Coxa magna occurs as a response to the hyperaemia around the hip subsequent to the synovitis, and also partly due to flattening of the epiphysis and broadening of the metaphysis.Which of these changes in the proximal femur can be prevented or minimised by treatment ?
Premature fusion of the capital femoral growth plate can neither be prevented nor even predicted and hence, coxa breva cannot be prevented. However, the “overgrowth” of the trochanter can possibly be prevented. Coxa magna again cannot be prevented but it may be possible to minimise the extent of coxa magna. Since femoral head extrusion appears to be the cause of the femoral head deformity, it follows that prevention of extrusion may prevent it from deformation.The goals of treatment of Perthes’ disease should be: i) to attempt to retain sphericity of the femoral head, ii) to minimise the extent of enlargement of the femoral head and iii) to prevent overgrowth of the greater trochanter.Prognostic factors in Perthes’ disease
Stulberg & Salter 8 noted that the only two long term prognostic features which determined the development of osteoarthrosis following Perthes’ disease were the age at the onset of the disease and the shape of the femoral head at the time of revascularisation. They also identified the short term prognostic features which determined the shape of the femoral head at the time of healing and they were; the age at onset of the disease, the extent of epiphyseal avascularity, persistent loss of hip movement and epiphyseal extrusion.
A closer look at these prognostic variables shows that the only prognostic factor related to femoral head deformity which is amenable to modulation by treatment is epiphyseal extrusion. Treatment should therefore, primarily be directed to preventing or reversing epiphyseal extrusion.
What are the currently available treatment options?
Treatment aimed at preventing femoral head deformity
The two main methods of treatment of Perthes’ disease aimed at preventing femoral head deformity are
- Weight relief
It is important to note that irrespective of the method adopted, treatment should continue till the avascular bone has been replaced by mature lamellar bone. Thus treatment should only cease after Stage III of the disease (Elizabethtown classification2 ). It is also emphasised that treatment is likely to be ineffective if offered in the stage of revascularisation (Stage III) as the biologically plastic bone has already been subjected to deforming stresses.
Weight relief is based on the belief that avoidance of weight-bearing stresses across the acetabular margin prevents femoral head deformation.
Prolonged weight relief in bed is not practiced any more, but weight-relief in the ambulant child is recommended by some surgeons. There is a suggestion that a combination of weight-relief and containment gives better results than containment alone6.
Containment entails placement of the anterolateral part of the femoral head within the confines of the acetabulum to prevent deformation by pressure from the acetabular rim. The anterolateral part of the femoral head can be positioned into the acetabulum by abducting and internally rotating the hip or by abducting and flexing the hip. This can be achieved either with the use of an orthosis which holds the hip in the appropriate position or by means of surgery.Containment by orthosis
Several braces which hold the hip abducted and flexed or abducted and internally rotated have been tried with some measure of success. The appliance needs to be worn constantly until revascularisation of the epiphysis is complete and this may take upto two years. The advantage this methods offers is the avoidance of surgery but excellent patient compliance is essential for it to work. Socio-economic factors may render this form of treatment inappropriate. Frequent radiological monitoring is essential to determine when to abandon the brace.Surgical containment
Containment can be achieved surgically either by performing a femoral subtrochanteric osteotomy (varus derotation or a varus extension osteotomy) or an innominate osteotomy.
A) Femoral osteotomy : The femoral osteotomy is designed to ensure that the part of the femur proximal to the osteotomy is held abducted and internally rotated or abducted and flexed to achieve containment. The procedure is simple, but some shortening of the femur may persist and a second operation is required to remove the implant which holds the osteotomy.
B) Innominate osteotomy : The orientation of the acetabulum is altered so as to cover the anterolateral aspect of the femoral epiphysis. The Salter’s osteotomy is the most widely used and good results have been reported with this procedure2 . The innominate osteotomy is technically more demanding than a femoral osteotomy and there is the suggestion that the procedure may cause an increase in the pressure on the femoral head.Treatment aimed at minimising greater trochanteric overgrowth
As the greater trochanter continues to grow when premature fusion of the capital femoral epiphysis occurs, arresting the growth of the greater trochanter would avoid its overgrowth with a resultant Trendelenburg gait. The use of trochanteric epiphysiodesis as part of the treatment of Perthes’ disease is not widely reported. Langenskiold4 recommended it; but it is only recently that authors have advocated its routine use3,5.
Surgery is only justifiable in children who present early in the course of the disease. The choice of the type of surgery ( femoral or innominate osteotomy ) depends solely on the preference of the surgeon. The author prefers to perform the femoral osteotomy because the femoral procedure is simpler and in addition, the trochanteric growth plate arrest can be performed simultaneously. A protocol of treatment is outlined in Table I.Contraindications for surgery
It is emphasised that surgery is not offered to children with Catterall group I involvement where only about one fourth of the epiphysis is avascular. In these chil- dren the prognosis is usually so good that surgery is seldom, if ever, justified. Surgery is also denied to patients who present in Stage III and Stage IV of the disease where the new bone on the lateral part of the epiphysis has already been subjected to deforming stresses. Surgery would not appreciably alter the final outcome in theseTABLE I - Indications for surgical containment in Perthes’ disease
Stage of the disease Under 7 years of Age Over 7 years of Age Elizabethtown Catterall's Group Catterall's Group Classification I II III IV I II III IV I - +/- +/- +/- - + + + II - +/- +/- +/- - + + + III - - - - - - - -
late cases. It is also important to avoid surgery in children who have stiff hips. If after a two week period of skin traction, the range of movement of the hip is not restored, surgery is to be avoided. Finally, surgery is not performed on children who are over 12 years of age at the onset of the disease. This is because the natural history of this adolescent form of Perthes’ disease is very different from that seen in younger children and containment surgery does not seem to offer any benefit in these patients.Pre-operative planning for surgery
Good quality antero-posterior radiographs of the pelvis showing both hips and the Lauenstein lateral projections are needed to enable the surgeon to determine the extent of epiphyseal involvement ( Catterall’s Group ), the stage of the disease and to determine whether there is any epiphyseal extrusion. Patients who are potential candidates for surgery based on the criteria listed in Table I are put to bed with skin traction for a period of two weeks. At the end of this period, the range of movement of the hips is evaluated. If all the movements have been restored, the patient would be offered a femoral varus derotation osteotomy. If internal rotation alone remains restricted, the patients would be offered a varus extension osteotomy. If an innominate osteotomy is preferred, it may be performed in both these situations.The surgical technique
The techniques described in the literature for the femoral osteotomy vary to some extent in technical details regarding the choice of the fixation device, the precise level of the osteotomy and the post-operative management. The author has been performing the femoral osteotomy in the following method for the last ten years with gratifying results.
A subtrochanteric open-wedge osteotomy is performed and the fragments are fixed with a plate pre-bent to 20 degrees. No post-operative plaster immobilisation is used. In addition to the varus osteotomy, the author routinely performs a trochanteric apophysiodesis. This is done by removing the flare of the trochanter and drilling the growth plate from its lateral aspect. The screws for fixing the plate are placed in such a manner that the first screw transfixes the trochanteric growth plate and the second screw is distal to the growth plate. The implants are retained for at least 18 months, during which time, the first two screws and the plate together act like a staple holding the lateral aspect of the growth plate. This additional precaution is taken in case the drilling of the growth plate has been inadequate for arresting the growth of the trochanter.What are the results of treatment ?
On analysing the short-term outcome of surgical treatment of children with Perthes’ disease according to the protocol outlined here, it was noted that sphericity of the femoral head was maintained in 62.5% of the operated children3 . The results were far poorer in children who had not received treatment (20% only had good spherical femoral heads). The children who had undergone surgery had less pronounced coxa magna than those who did not undergo surgery. It is clear from this study that the frequency of coxa irregularis and coxa magna can be minimised by early surgery. It is also important to note that in this study, 60% of children who did not receive treatment had distinctly poor results as opposed to less than 15% of poor results among those who were operated.
In carefully selected cases, surgical containment, if performed early in the course of the disease, significantly reduces the risk of coxa magna and coxa irregularis.
Thus, the morphological changes in the hip which predispose to osteoarthrosis may be either prevented or minimised by containment.
What are the limitations of treatment ?
A small but significant proportion of children with Perthes’
disease in India, are not candidates for surgical containment as outlined in this review. They either present too late in the course of the disease with established deformation of the femoral head or are older than 12 years of age. The frequency of late presentation can be minimised by an increase in the awareness among orthopaedic surgeons that it is important to ensure containment early in the course of the disease. The treatment options for those older than 12 years of age at the onset of the disease remain unclear at present as the disease appears to behave very differently and often unpredictably in the older child.
Over the years a clearer understanding of Perthes’ disease has led to vastly improved results in treatment. It is hoped that more information will be gathered in the years to come to enable us to treat this enigmatic disease even more effectively in the future.References:
- Bajaj HN, Joseph B, Chacko V. Radiological changes in late untreated Perthes’ disease. Ind J Radiology & Imaging 1988; 42: 365-8.
- Canale ST, D’Anca AF, Cotler JM, Snedden HE. Innominate osteotomy in Legg-Calve-Perthes’ disease. J Bone Joint Surg (Am) 1972; 54-A: 25-40.
- Joseph B, Srinivas G, Thomas R. Management of Perthes’ disease of late onset in southern India. The evaluation of a surgical method. J Bone Joint Surg (Br) 1996; 78-B: 625-30.
- Langenskiold A. Changes in the capital growth plate and the proximal femoral metaphysis in Legg-Calve-Perthes’ disease. Clin Orthop 1980; 150: 110-4.
- Matan AJ, Stevens PM, Smith JT, Santora SD. Combination trochanteric arrest and intertrochanteric osteotomy for Perthes’ disease. J Pediatr Orthop 1996; 16: 10-4.
- Mintowt-Czyz W, Tayton K. Indication for weight relief and containment in the treatment of Perthes’ disease. Acta Orthop Scand 1983; 54: 439-45.
- Muirhead-Allwood W, Catterall A. The treatment of Perthes’ disease: the results of a trial of treatment. J Bone Joint Surg (Br) 1982; 64-B: 282-5.
- Stulberg SD, Salter RB. The natural course of Legg- Perthes’ disease and its relationship to degenerative arthritis of the hip. A long-term follow-up study. Orthop Trans 1977; 1: 105-6.
Is arthrodesis of the shoulder justifiable in young children ?
Arthrodesis of the shoulder may be considered as an option for restoring stability of the shoulder in children when the shoulder is paralysed following poliomyelitis or obstetric paralysis. However, the classical teaching has been that shoulder arthrodesis be deferred till skeletal maturity. Is it essential to wait till the child is skeletally mature before performing this procedure
The three main objections to performing a shoulder arthrodesis in young children are :
- 1. Proximal humeral physeal damage may occur with arrest of longitudunal growth of the humerus.
- 2. The femoral head and the glenoid are small in children and consequently it may be difficult to achieve fusion.
- 3. With growth, the initial position of fusion may alter and this may compromise the long-term result.
The proximal humeral physis contributes 80% of the longitudinal growth of the humerus and since the physis is so close to the shoulder it is possible that it may get damaged during the process of denuding the humeral head of articular cartilage. The internal fixation device used to stabilise the arthrodesis could also contribute to the damage to the physis. However, based on available literature, there does not appear to be a great deal of evidence to support the view that physeal damage and consequent shortening of the humerus occurs frequently. Among papers devoted to the role of shoulder arthrodesis in children, two specifically address the issue of shortening of the humerus at skeletal maturity. Data from these papers shown below suggest that some humeral shortening does occur. However, the degree of shortening was not of an alarming magnitude and it is unclear whether some degree of shortening secondary to the paralysis had been present prior to arthrodesis.
Author & year No. of cases Follow-up Mean shortening Makin 1977 7 children 10.1 years 2.3 cm Mah & Hall 1990 10 children 15.1 years 3.4 cm III Yes Yes Human position
Failure to achieve sound fusion
De Velasco Polo & Monterrubio (1973) reported a pseudarthrosis rate of 20% in a series of 31 children. However, the technique of internal fixation employed was a single Steinmann pin and a pure intra-articular arthrodesis was performed. Subsequent reports have recorded a much higher union rate when more appropriate surgical techniques were employed, as shown below.
Author & year No. of cases No. fused Technique of fusion Internal fixation Makin 1977 7 7 Intra-articular 2 Steinmann pins White et al 1989 5 5 Intra + extra-articular Threaded pins Mah & Hall 1990 10 10 Intra + extra-articular Screws Pruitt et al 1992 17 16 Intra + extra-articular Hagie pins/plate & screws
Thus it appears that it is possible to achieve fusion of the shoulder in children, in the majority of instances.Loss of position of fusion
Saha (1967) had suggested that a downward bending of the arm at the epiphyseal plate can occur with gradual loss of the position of abduction even after sound fusion of the shoulder had been achieved. White et al (1989) reported a series of five cases of shoulder arthrodesis in children and noted that the angle of abduction decreased progressively in four cases. All the four children had fusions with 30o to 40o of abduction and this had reduced to less than 10o within a period of 12 months after the operation. Based on their experience the authors recommended that shoulder arthrodesis be deferred till skeletal maturity.
Makin (1977) arthrodesed the shoulders of seven children with the arm in wide abduction of almost 90o, in order to compensate for the possible loss of the abducted position of fusion with time. However, the position of fusion did not alter in these children. Consequently, at skeletal maturity, these patients were unable to bring their arms down to the side of the body. They also had winging of the scapula which was cosmetically unacceptable. Mah & Hall (1990) also did not encounter any appreciable loss of postion of fusion in their series of 10 children.
From the cases reported in literature, it appears that in the majority of instances, there is no loss of position of fusion.The three theoretical objections to performing shoulder arthrodesis in children are not supported by sufficiently convincing evidence in the literature. In the light of this review it may be worthwhile, considering Makin's arguments for performing arthrodesis of the shoulder in children over seven years of age:
" The importance of early rehabilitation, particularly of the upper limbs, is recognised by all who deal with children. Prostheses are fitted as early as feasible so that the growing child has an upper extremity which is put to maximum use at as early an age as possible. Similarly a flail shoulder should be eliminated as soon as this can be done to allow maximum use of the extremity."
Another controversial issue is the optimal position of arthrodesis. It would be ideal if, after the arthrodesis:
- 1. the arm can rest at the side of the trunk,
- 2. there is no marked winging of the scapula and
- 3. the arm can be raised well above the head.
Recommendations in the literature regarding the position of fusion required to achieve these goals have varied from :
45° to 90° of abduction
15° to 50° of forward flexion
45° external rotation to 45° of internal rotation.
Pruitt et al (1992) on reviewing the results of shoulder arthrodesis in 17 children, noted that the position of fusion varies considerably from patient to patient ( ranges : 40o to 70o abduction, 10o to 55o flexion, 20o external rotation to 30o internal rotation). Despite this wide variation in the position of fusion the authors reported satisfactory results in all the patients both in terms of function and patient satisfaction. They went on to suggest that the exact position of fusion is not as important as the stability gained at the glenohumeral joint. However, they suggest that excessive abduction and flexion should be avoided to prevent cosmetically unacceptable winging of the scapula. Mah & Hall (1990) also noted the lack of consensus regarding the position of fusion but went on to recommend that in children the shoulder should be fused in 45o abduction, 25o flexion and 25o internal rotation.
Finally, those who perform shoulder arthrodesis in young children should carefully document the immediate post-operative position of fusion and monitor the patients over several years to evaluate whether loss of position does occur. If loss of abduction does occur, an osteotomy of the proximal humerus could be done without too much difficulty.References:
- Makin M. Early arthrodesis for a flail shoulder in young children. J Bone Joint Surg (Am) 1977;59:317-21.
- Mah JY, Hall JE. Arthrodesis of the shoulder in children. J Bone Joint Surg (Am) 1990;72:582-6.
- DeVelasco Polo G, Monterrubio AC. Arthrodesis of the shoulder. Clin Orthop 1973;90:178-82.
- White JI. Hoffer MM, Lehman M. Arthrodesis of the paralytic shoulder. J Pediatr Orthop 1989;9:684-6.
- Pruitt DL, Hulsey RE, Fink B, Manske PR. Shoulder arthrodesis in pediatric patients. J Pediatr Orthop 1992;12:640-5.
- Saha AK. Surgery of the paralysed and flail shoulder. Acta Orthop Scand (Suppl) 1967;97:13-4.
Published for and on behalf of the Paediatric Orthopaedic Society of India byDr.Benjamin Joseph
Department of Orthopaedics Dr. K.Sriram
Kasturba Medical College & Hospital Dr. Benjamin Joseph
MANIPAL 576 119 , Karnataka State
Fellowship post in Paediatric Orthopaedics
The Department of Orthopaedics at Kasturba Medical College invites applications for a one year Fellowship in Paediatric Orthopaedics. The Fellowship offers an exposure to a broad range of Paediatric Orthopaedic diseases. The fellow will receive a monthly stipend and a certificate from the Manipal Academy of Higher Education ( A Deemed University ) on completion of the Fellowship.
For further information contact : Dr.Benjamin Joseph, Paediatric Orthopaedic Service, Department of Orthopaedics, Kasturba Hospital, Manipal 576 119, Karnataka. Fax. 08252 70062. Last date for application: 15 March 1998.
Publications from the members of POSI
The Editors welcome copies of papers published by members of POSI for inclusion in the abstract section of POSITIVE. It would be a good way to keep fellow members informed about our contributions in the field of Paediatric Orthopaedics.Formation of working groups for common paediatric orthopaedic problems
It is proposed to form working groups to share information and plan multicentre studies on common problems we see in our paediatric orthopaedic practice. The modalities of the functioning of these working groups would be discussed in the 1998 meeting of POSI. To begin with, study groups on clubfoot, developmental dysplasia of the hip and Perthes' disease are planned. Those interested in joining these groups kindly send in your names to Dr. Ashok Johari.