Introduction

About 1920, acetabular roof plasty began to attract wider attention. Open repositions of congenital hip dislocations did not give good results. Acetabular roof plasty, on the other hand, made it possible to create a bone support for the subluxed femoral head, which tended to protrude.
The idea of acetabular roof plasty was given by Franz Konig in 1891. He performed it on two girls who, however, died of accidental infectious diseases. Konig’s idea was taken up by Fairbank (1911), and later by other authors such as Albee (1915), Spitzy (1924), M. Lance (1925), Ghormley (1931), Schede (1933), and Gili (1935).
In Poland, the subject of acetabular roof plasty was discussed at the Congress of the Polish Society of Orthopaedics and Traumatology in 1931 in Warsaw. There I presented the results of 10 acetabular roof plasty procedures performed between 1927 and 1930, including 5 plasty procedures in children due to subluxation and 1 plasty due to hip dislocation. The rest was performed in adults.
The results of these procedures led me to the conclusion that an important part of roof plasty should not involve a bone graft as support, but lowering of the acetabular roof itself so that it covers the femoral head, with the graft filling the gap that arises above the lowered roof and stimulating acetabular roof growth. I expressed this opinion at the Congress of the Italian Orthopaedic Society in 1932 in Bologna. I emphasized there that “the incision with the chisel should be deep, parallel to the upper surface of the articular acetabular roof, and should cover the entire upper half of the acetabulum, not excluding the posterior part”. “The incised part of the roof must be lowered easily, and with it the femoral head.” “The lowering of the roof should therefore also remove the subluxation of the head.”
Connecting the roof plasty with simultaneous subluxation reduction by lowering the acetabular roof was my intention in roof plasty, which I continued to systematically perform.
Unfortunately, only the documentation of the published cases (1931, 1932, 1933) from the pre-war period have been preserved.
After becoming the head of the Orthopaedic Department of the University of Poznań in 1945, I continued to improve the method used to reconstruct the acetabular roof with the simultaneous reduction of the femoral head (Fig. 1). In 1958 I called this method – that was initially nameless – a semi-circular supra-acetabular osteotomy or transiliac osteotomy. I presented it in a discussion at the meeting of the Czech Orthopaedic Society in Prague in 1959.
W. D. Czaklin appreciated this method and included it, providing the year 1959, (pages 403 and 404, Fig. 288a) in his textbook “Frames of Surgical Orthopaedics and Traumatology”, Moscow 1964.
Over the years having developed the concept of the reconstruction of the acetabular roof with simultaneous reduction, I did not yet know Salter’s method or Pemberton’s method, from which my procedure differs in technical details. Salter announced his method in 1961, Pemberton in 1953, and in 1965 (in a script in 1961).
In 1953, in our Department, we were fascinated by the acetabular plasty method according to Colonna. We started to systematically use this method, supplementing it with the shortening of the shaft and correcting the position of the femoral neck via osteotomy (detorsion, devalgisation according to ZahradniČek).
The combination of Colonna’s method and ZahradniČek’s method, which we called hip reduction-reconstruction, seemed logical and very convincing to us. Between 1953 and 1965 we operated on 1,044 hips in the Department using this method. However, our follow-up studies of the late results of using this method showed that they are not as good as we expected. Colonna’s acetabular plasty in subluxation cases had particularly unfavourable results. This prompted us to revert to transiliac osteotomy for both subluxation and hip dislocation.

Surgical technique

As in the method of reconstruction-reduction, we adopted the principle of removing all defects of the dysplastic hip during one surgical session, namely removing the femoral head displacement, obtaining the correct position of the head in the bottom of the acetabulum, and giving the upper part of the acetabulum as normal shape as possible.
Before the procedure, all hip defects to be removed should be determined in detail. Therefore, it is necessary to perform an X-ray to determine value of the angles of antetorsions and to check whether and in what position of the thigh the femoral head can penetrate the acetabulum deep enough or to determine whether the existence of soft tissue interposition between the head and the bottom of the acetabulum should be taken into consideration.
These examinations allow for the appropriate treatment option to be selected. The variants include:
1. transiliac osteotomy without joint opening:
a) without correction of the femoral neck,
b) with correction of femoral neck anteversion and deformity of the femoral neck;
2. transiliac osteotomy with joint opening and head subluxation reduction;
3. transiliac osteotomy with reduction of dislocation and shortening of the femoral shaft, with correction of femoral neck anteversion and valgus deformity and shortening of the femoral shaft.
In all variants, the most important element is added, i.e. reduction of the subluxation in respect to the dislocation.
Skin incision (Fig. 2). Place the child on his/her opposite side for the procedure. Skin incision is common to all variants of the procedure, which allows for any extension of the procedure, if it turns out to be necessary during the procedure. The incision begins at the edge of the iliac crest about 3 cm posteriorly (dorsally) from the anterior superior iliac spine and runs towards the greater trochanter and continues downwards on the external-posterior surface of the thigh for about 10-15 cm (depending on the variant).
The femoral fascia is cut from the anterior superior iliac spine towards the greater trochanter and further as a cutaneous incision.
Access to the joint. First, we get to the joint from the front along the front edge of gluteus medius and sartorius. Tensor fasciae is often very thin. Then we move it sharply away from its fascia (we sew it later). To access the superior and posterior part of the acetabulum, we move the muscles bluntly and sharply away from the joint capsule covering the femoral head. These are gluteus medius and minimus, and short external rotators. For better access to the dorsal part of the femoral head and the acetabulum, cut the periosteum along the outer edge of the iliac crest, starting from spina iliaca anterior superior towards the back, approx. 3-4 cm. The removal of the periosteum together with the muscles should be carried out carefully and extend to the edge of foramen ischiadicum majus.

Transiliac osteotomy
It involves a proper chiselling of the iliac bone just above the acetabulum and modelling of the mobilized upper half of the acetabulum as a bone covering for the femoral head.
For this purpose, we make an incision in the periosteum along the outer edge of the iliac bone in the section between the inferior and superior iliac spine. We insert a raspatory between the periosteum and the pelvic side of the iliac bone towards foramen ischiadicum majus and move the periosteum away from the bone. From the outside, we introduce the Hohmann retractor in foramen ischiadicum, the opening of which we have already exposed.
Begin chiselling of the iliac bone just above the inferior anterior iliac spine. The chisel cuts the inner and outer cortical layers and cuts as parallel as possible to the semi-circular articular surface of the roof without disturbing it. The chisel stops near foramen ischiadicum without cutting its bony margin.
The incised roof forms a kind of flap consisting of a cartilage layer and an adjacent cortical layer of bone. It can now be easily lowered from the front and top, and the posterior edge of the acetabulum can be turned slightly forward. This is possible if the osteochondral flap is not too thick and has sufficient flexibility. Such flexibility decreases with age.
When chiselling the flap, the superior acetabular edge must not be disturbed under any circumstances, i.e. the place where the rise (horizontal) of the acetabular edge takes place. The chisel is placed above this edge.
After the roof has been lowered, a gap is created, which must be filled with bone grafts. In children whose wing of ilium is sufficiently thick, we cut bone wedges from the iliac crest in the vicinity. In children with a thin wing of ilium, we use freeze-dried hemogenic spongy bone from the bone bank. Freeze-dried bone heals well, only requires a little more time to reconstruct.
When closing the wound, one should pay attention to separate suturing of the periosteum together with gluteus medius and minimus to the bony edge of the wing of ilium and separate suturing of the fascia. Failure to sew the muscles in their proper tension to the bone weakens them, and they play an important role in stabilizing the pelvis (Trendelenhurg sign). We close the rest of the wound as usual.

Transiliac osteotomy with reduction of displaced femoral head without capsule opening
In young children, after separating the muscles from the joint capsule, it is sometimes possible to reduce the displaced femoral head into the acetabulum by simply pressing with the thumb through the capsule without opening it. If there is no coexistence of increased femoral neck anteversion or valgus deformity, one can confine to supplementing this reposition with transiliac osteotomy
(Fig. 3).
Attention should be paid to the precise separation of the capsule from the wall of the secondary acetabulum, where the union of the capsule with the bone is very tight. The capsule should be separated up to the edge of the primary acetabulum. Sometimes, after repositioning, there is an excess capsule that forms a thick fold at the front. This fold can be cut out and the capsule can be sewn up.
The variant of isolated transiliac osteotomy is performed very rarely. In most cases, acetabular dysplasia is combined with femoral neck anteversion and valgus deformity.
Transiliac osteotomy with simultaneous directional osteotomy of the femur
Counting on the fact that a well-shaped acetabular roof would normalize the femoral neck angles over time may fail. It is safer to correct the position of the femoral neck during the procedure.
In these cases, we extend the incision downwards. We reveal vastus lateralis and cross its attachment transversely below the greater trochanter. With the second incision along its dorsal edge, we cut the periosteum and move the muscles together with the periosteum from the femur in the subtrochanteric part. Pre-operative radiological calculations of the angles of femoral neck anteversion and valgus deformity are verified by eye.
Correction must be precise. We mark the femoral neck axis with the directional Kirschner wire and control the position of the wire with an X-ray image during the procedure. In order to mark the degree of femoral neck anteversion, we insert short Kirschner wires as markers vertically into the femur above and below the osteotomy site. We cut a properly shaped bone wedge from the femur.
For the fixation of the fragments, we use a metal angular plate, the size of which is selected according to the preoperative X-ray. The tip of the plate is inserted into the neck along the directional Kirschner wire. First, the plate is temporarily attached to the shaft with a bone-holder after the neck angles have been corrected. We put the head into the acetabulum and check if it is not pressed against the acetabulum by shortened muscles, if the hip joint is stable in the zero position and if a certain degree of inward rotation is preserved after derotation. If it is absent, we reduce the degree of correction of femoral anteversion. After making additional corrections, we attach the plate with screws to the shaft of the femur and sew vastus lateralis in its place. Care should be taken to suture the periosteum and secure tendons of gluteus maximus in the right place.
As for the sequence of procedures, we first perform directional osteotomy on the femur, suture the muscles on the femur, and then perform transiliac osteotomy. To fill the gap after lowering the roof, we use a bone wedge cut from the femur during osteotomy.

Transiliac osteotomy with opening
of the joint capsule and subluxation reduction
When interposition is suspected, we open the joint by cutting the front part of the capsule crosswise along the edge of the primary acetabulum.
Before opening it, bluntly separate rectus femoris and iliopsoas from the capsule. Near the anterior inferior iliac spine, tendo reflexum m. recti departs from the rectus muscle. In cases of subluxation, it often forms a wide tendinous cap above the femoral head that fuses with the capsule and strengthens it. We cut this tendon off the capsule to be sewn back to it again in the right place as its tensioner.
Lig. teres – if preserved – is usually thick and elongated. After it is cut and the head is balanced out of the acetabulum, we get a good view of the acetabulum. We empty it from all interposed soft parts, so we cut out lig. teres, lig. transversum, a curled helix that is sometimes membranous and obscures the acetabulum. We leave the acetabular fat lining (pulvinar).
Opening the joint allows you to see the type of defect in the acetabular roof. Very often we find a defect in the upper anterior edge of the acetabulum and the anteversion of the femoral head in relation to the acetabulum. These are the cases in which – during conservative treatment – the thighs were held in flexion and hyper abduction for a long time (hyperabductio – Fig. 4).
Performing transiliac osteotomy with an open joint is much easier, because seeing the inside of the acetabulum, you can precisely guide the chisel. Corrective osteotomy of the femur – if necessary – can also be performed without an X-ray, as the head and neck of the thigh are visible to the eye.
The procedures are performed in the following order: first arthrotomy and emptying of the acetabulum, then directional osteotomy of the femur with fixation of the fragments using an angular plate and with the suturing of the muscles to the thigh. As the last step, transiliac osteotomy is performed.
Transiliac osteotomy with hip dislocation reduction
The technical performance of the procedure is similar to that of subluxation, but with some differences. After the joint capsule is completely dissected from the surrounding muscles, also in the dorsal and anteroposterior parts (rectus and iliopsoas), the entire joint capsule is excised, leaving only a wide collar at the base of the neck (circulus arteriosus!).
In significant dislocations, the access to the acetabulum is often obscured by overgrown iliopsoas. In that case, we cut its tendinous part obliquely, which allows to move it away from the bottom of the acetabulum and empty the acetabulum from the soft parts filling it.
We combine directional femoral osteotomy with shaft shortening by cutting out a piece of the shaft. The length of the excised section depends on the degree of dislocation. The higher the dislocated femoral head, the longer the shaft section resected must be. The test for sufficient shortening of the shaft is the lack of support of the head to the acetabulum when trying to insert it into the emptied acetabulum. When assessing the support, we must take into account that the lowering of the roof after transiliac osteotomy exerts some pressure from above on the femoral head.

Post-operative procedure
The child is given a spica cast. After 3-4 weeks, we cut out the front part of the pelvic part and the child can sit down. About 6 weeks after surgery, we remove the plaster cast completely. The child remains in bed and can move his/her legs freely and sit down. The child must not kneel or get up. For the night, the child receives a plaster half-cast for the pelvis and both legs slightly abducted.
We do not rush to load the operated joint. We wait for a while after the reconstruction of the acetabular roof and the bone grafts placed there, as well as the union of the femur at the site of directional osteotomy. X-ray imaging, the child’s age and the type of procedure will show when the operated limb can be loaded. In general, we do not allow the child to walk for 3-4 months after surgery. Disciplined children are allowed to walk using crutches without putting load on the limb. In children with bilateral hip dysplasia, we wait about 4-6 weeks to perform surgery of the other side.

Indications
Dysplasia of the acetabular roof, especially its anterior superior segment, is considered an indication for transiliac osteotomy. It may be the only hip defect. Most often, however, it is associated with other symptoms of dysplasia, such as increased anteversion or valgus deformity of the formal neck, which may be accompanied by head displacements, either partial (subluxations) or complete (dislocations). A large percentage of our operated children had residual dysplasia after completed conservative treatment.
The optimal age for transiliac osteotomy is 3 years, especially for children who need to have directional femoral osteotomy performed that requires the use of a metal plate for fracture fixation. The upper age limit is determined by the flexibility of the cartilage part of the chiselled acetabular roof. Such flexibility decreases with age. In a child over 6 years of age, there may be difficulties with modelling this excerpt.

Discussion
Colonna’s acetabuloplasty forms a new acetabulum. This method’s advantage is that the acetabular shape can completely adapt to the shape and size of the head, provided that the acetabular roof is sufficiently thick. However, if there is a deep defect in the roof, as seen in marginal subluxations, acetabuloplasty may lack the top bony covering of the head. In transiliac osteotomy, it is always possible to obtain good head coverage from the top, front and back.
In acetabuloplasty, the head is separated from the bare bone in the acetabulum only by a layer of the capsule, which turns out to be a biologically insufficient interposition. Hence the early appearance of pre-arthrosis after acetabuloplasty. In transiliac osteotomy, the head is covered with acetabular cartilage, hence the hope that the operated joint will retain its function longer.
In Pemberton’s surgery the osteochondral flap is thick and the line of the iliac chiselling is directed to the Y-shaped cartilage, and here is the pivot point of the osteochondral flap in the roof. In transiliac osteotomy, the pivot point lies in the thin bone peduncle close to foramen obturatum. As a result, the osteochondral flap can not only lower, but also make a certain rotation so that the posterior edge of the acetabulum can move a little forward. According to Salter’s method, the pivot point lies in the interpubic joint. It is alleged that the rotation of the entire acetabulum together with a part of the pelvic ring lengthens the limb and makes the posterior edge of the acetabulum shallow. Salter’s method is technically more complex than transiliac osteotomy.
The latter can be accused of the fact that in the case of a narrow acetabulum, the upper part of which has the shape of a gothic ogival (ogival acetabulum), there is no congruence between the head and the roof covering it, and that the ogival roof pushes it down. Indeed, in some cases we observe a low head position after surgery. Sometimes it is the fault of the overzealous lowering of the roof by the surgeon, but the reason may also lie in the ogival shape of the roof. In many cases, a low head position corrects itself. However, this is a detail that requires further observation.
Overall, it seems to us that the reposition of hip subluxations and dislocations in combination with transiliac osteotomy gives better results than previously achieved. A meaningful evaluation will depend on the long-term results.

References
1. Albee FH: New York Med. J., 1915, t. 102, p. 21-33.
2. Czaklin WD.: Osnowy operatywnoi ortopedii i trawmatołogii. Medicina, Moskwa, 1964.
3. Dega W: Chir. Narzadow Ruchu Ortop. Pol., 1929, t. 2, p. 221-253.
4. Dega W: Chir. Narzadow Ruchu. Ortop. Pol., 1931, t. 4, p. 533-591.
5. Dega W: Atti del XXIX Congr. Soc. Italiana di Ortop. Bologna, 15-17.10.1932.
6. Dega W: Chir. Narzadow Ruchu Ortop. Pol., 1938, t. VI, p. 721-733.
7. Dega W: Pfannendach- piastik. G. Thieme, Stuttgart, 1965.
8. Dega W: Arch. Orthop. Unfall-Chir., 1966, t. 60, p. 16-29.
9. Dega W: Verh. Deutsch. Ges. Orthop. u. Traum. 56 Kongress, Wien, 1969.
10. Dega W, Król J, Polakowski L: J. Bone .Tt Sur- gery, 1959, t. 41-A, p. ???
11. Dega W: Beitr. Orthop. Traum., 1964, t. 2, p. 642.
12. Fairbank HAI: Proc. Royal Soc. Medecine, 1923, t. 16, p. 15-23.
13. Ghormley RK: J. Bone Jt Surgery, 1931, t. 13, p. 784.
14. Gili AB: J. Bone Jt Surgery, 1935, t. 173, p. 48.
15. Konig Fr: Zbl. Chir., 1891, 18 Beilage, p. 46.
16. Lance M: Presse Med., 1925, t. 56.
17. Lance M: Buli. et Mem. de la Soc. Nat. de Chir., 1927, t. 53, p. 1.
18. Pemberton PA: J. Bonę Jt Surgery, 1965, t. 47-A, p. 65.
19. Salter RB: J. Bone Jt Surgery, 1961, t. 43-B, p. 518.
20. Salter RB: J. Bone Jt Surgery, 1966, t. 48-A, p. 1413.
21. Schede Fr: Zeitschr. Orthop., 1933, t. 68, (Verhandlungsbd.).
22. Spitzy H: Zeitschr. f. Orthop. Chir., 1923, t. 54.
23. Spitzy H: Zentrbl. f. Chir., 1928, t. 21.