Reconstruction of the hip joint in children after septic coxitis

Cover Page



Surgical treatment of destructive dislocations is problematic, and controversial. Treatment options range from active monitoring to complex multistage reconstruction protocols.

Materials and methods.

The results of treatment of 17 children with disruptive dislocation of the hip (class IV according to the Choi classification) are presented. The mean age at which treatment was started was 6.3 ± 0.5 years (range: 3-9 years). Treatment strategies included formation of the weight-bearing contact of the proximal femur and acetabulum and ilium osteotomy using an Ilizarov frame. The follow-up period lasted from 2 to 11 years. Clinical results were assessed using Colton’s criteria. Assessment of the radiology findings included the anatomy relations and congruence of the articular surfaces. The patients were divided into two groups, according to age, for data analysis.


The majority (88%) of the children had positive anatomical and functional outcomes. The functional results were good (12-15 points), satisfactory (9-11 points), and unsatisfactory (8 points) for 3, 5, and 1, respectively, among 3-6 year olds and 2, 6, and 0, respectively, among 7-9 year olds. The distribution of joint congruence types I, III, and IV was 2, 4, and 3, respectively, in the younger group and 1, 6, and 1, respectively, in the older group.


The use of reconstructive surgery using the Ilizarov apparatus in children with a defect of the proximal femur creates conditions for the adaptation of incongruent articular components without causing decompensation of the joint.

Full Text


Septic coxitis is a relatively rare condition that requires urgent attention owing to its severe complications [1]. The complications are quite diverse and may manifest as growth plate injury, deformity, and destruction of articular components, and incorrect joint relations [2, 3]. These changes significantly affect limb function; therefore, one of the main treatment objectives is to increase the ability to support one’s weight [4, 5]. Treatment depends on the degree of destructive changes. It involves, as a rule, correction of the proximal femur deformity, restoration of lost femur elements, and restoration and stabilization of the articular relations [6]. The greatest difficulty is the rehabilitation of patients with proximal femur defects [7]. Treatment options range from dynamic monitoring to complex multistage reconstructive surgeries [8]. There is a standard regarding the nature and extent of surgical interventions among the surgical treatment advocates [5]. The prevailing view is that reconstructive surgeries, aimed at the formation of the proximal femur and restoration of joint relations, are only possible in young children [4, 9]. Palliative treatment is recommended for children over the age of 6 years [10]. At the Ilizarov Research Center for Restorative Traumatology and Orthopedics, Russian Federation, reconstructive techniques with the use of external fixation devices are used to treat children up to 10 years of age. This study aimed at analyzing the immediate and midterm results of treatment of preschool and early school-age children with defects of the femoral head and neck secondary to septic coxitis.

Materials and methods

Treatment outcomes of 17 children with septic coxitis complications were analyzed. Patients were treated at the Ilizarov Center between 2002 and 2011. Parents or legal guardians of all patients signed a voluntary informed consent form to participate in the study and undergo surgery (protocol of the meeting of the Ethical Committee of the Ilizarov Center no. 2, 03.31.2016). All patients were diagnosed with coxitis within the first 3 months of life. There were four cases of knee joint disorder. The distribution by sex was nine girls and eight boys. The average age at the beginning of treatment was 6.3 ± 0.5 years (3-9 years).

Patient inclusion criteria were as follows: destructive hip dislocation corresponding to class IV according to the Choi classification [2], aged less than 10 years, and undergoing at least 2 years of follow-up. Patients with distention dislocation of the hip joint were excluded from this study.

Treatment outcomes were analyzed from 2 to 11 years. Clinical results were evaluated according to the Colton criteria [11]. Radiological data were obtained from pelvic radiographs taken in anteroposterior projection. Joint pathology severity was determined according to the Choi classification, condition of the acetabulum (acetabular index, AI). Support surface inclination angle (WBS) and acetabular coefficient (AC) were evaluated [12]. Measurements were determined by radiographs taken before treatment, after removal of the Ilizarov apparatus, and on the last follow-up visit.

On the radiographs taken after treatment, the following indicators of the joint relations were determined:

Lateral displacement angle (LDA): the angle between the vertical line that goes through the tear drop and the tangent to the lower medial edge of the femoral neck. The vertex of the angle is positioned at the intersection of the vertical line and the perpendicular line drawn through the top pole of the femoral head (normal range 20°-25°) [12].

Degree of acetabular coverage (acetabular head index, AHI) [12].

Joint surface congruence according to Coleman (Type I, spherical congruent; Type II, spherical incongruent; Type III, aspherical congruent; and Type IV, aspherical incongruent) [13].

The results were considered good if the parameters were as follows: LDA ≤ 25°, AHI ≥ 80%, and there was Type I or III congruence. Results were considered satisfactory if the parameters were as follows: LDA ≤ 30°, AHI ≥ 70%, and Type III or IV congruence. Results were considered unsatisfactory if the parameters were as follows: LDA > 30°, AHI < 70%, and Type IV congruence.

This study was conducted in accordance with the standards laid down in the World Medical Association Declaration of Helsinki, amended by the Ministry of Health, Russian Federation. Legal representatives for all patients signed informed consent to participate in this research study, in which data were identified.

The treatment method involved the use of the Ilizarov apparatus, which included the formation of the load-bearing surface of the proximal femur, its insertion into the acetabulum, iliac osteotomy, and transposition of the acetabulum for stabilization of the achieved relations. Three variations of the technique were used for the femur reconstruction. In six cases with Type IVa congruence and a sufficient length of the femoral neck stump (no less than 25% of the length of the intact joint neck), we performed a C-shaped osteotomy through the base of the femoral neck and gradual transposition of fragments to form the femoral neck-shaft inclination angle.

In eight cases with IVa (femoral neck stump less than 25%) and Type IVb in children older than 4 years, the load-bearing surface of the femur was formed using double transtrochanteric osteotomy, and lateral transposition of the tip of the greater trochanter and the distal fragment of the femur.

In three younger children with Type IVa, the joint reconstruction was accomplished with valgus lateral intertrochanteric osteotomy. In nine cases, we performed closed reduction of the dislocation after formation of the proximal femur. In the rest of the cases, open reduction was required. Iliac osteotomy was performed 2 to 3 weeks after the dislocation reduction. The average duration of the treatment with the external fixation device ranged between 100 and 140 days. After removal of the external fixation, we applied a removable plaster cast. All patients underwent rehabilitation treatment. Full weight-bearing was permitted in 6 to 8 weeks after removal of the external fixation.


All patients were divided into two groups based on their age. The first group included nine children aged between 2 and 6 years. The second group included eight patients aged between 7 and 9 years.

Clinical signs of pathology in patients were as follows: pathologic gait, positive Trendelenburg sign and Dupuytren [piston, telescoping] test, restriction of retraction and extension in the hip joint, relative shortening of the limb (group 1 by 4.9 ± 0.4 cm; group 2 by 6 ± 0.4 cm). In group 1, the range of movement in the sagittal plane was 80-110°. In the frontal plane, it was 35-45°. In group 2, the mobility in the sagittal plane was 80-100° and 25-35° in the frontal plane. Moderate pain was present only in three patients in group 2.

Distribution by the type of joint deformity (Choi) in group 1 was as follows: six joints with Type IVa, tree with Type IVb. In group 2, it was as follows: five with Type IVa and three with Type IVb. There were no significant differences in the indicators of acetabular dysplasia between the groups, which were 35.6° ± 0.7° and 35.3° ± 0.9°, respectively.

Double transtrochanteric osteotomy was performed in three patients in group 1 and in five in group 2. Elongation of the femoral neck was performed in three cases in each group.

In group 1, we were able to achieve joint stability with closed reduction in only one case. In the majority of cases (seven cases), open reduction was required. Of these, in five cases, reduction was performed after the intervention on the femoral component. In two joints, open reduction was performed during the second surgery 2 to 3 months after removal of the external fixation, owing to the joint instability. In group 2, we performed closed reduction of dislocation in seven cases.

The analysis of functional results (see. Table 1) showed improved gait in most patients in group 1. In five cases, the Trendelenburg sign remained positive, mainly Type I. In all joints, the Dupuytren symptom was negative. In two cases, there was a 50% improvement in the joint range of motion compared with the range of motion preoperatively. In other cases, the range of motion improvemed over 75% compared to the baseline. The average joint mobility grade was 2.8 ± 0.2 points. In all cases, pain syndrome was absent or insignificant. In six patients, full compensation of the femur shortening was achieved. In three cases, there still was relative limb shortening of 2 to 3 cm. The distribution of functional results in group 1 (including reoperations) according to Colton criteria was as follows: good (12-15 points), three joints; satisfactory (9-11 points), five joints; and unsatisfactory (8 points), one joint.

In the second group the average indicators of gait and load bearing ability of the limb were lower than in group 1 (Table 1). The piston test was negative in all cases. Significantly limited mobility (50% of the initial) was present in only one patient. The average indicator of mobility was higher than in group 1. It should be noted, that these differences were not statistically significant. Insignificant pain syndrome was noted in two cases. In three cases, the relative shortening of the leg remained within the 3-cm range.

The functional results of group 2 were as follows: good (12-14 points) results in 2 joints and satisfactory (9-11) in 6 joints. Good results were achieved only in children with Type IVa joints.

Radiological data analysis after surgery demonstrated a significant improvement in the acetabulum, that is, decreased acetabular index and increased acetabular coefficient. In group 1, there was a further increase of AC in the course of dynamic observation, which indicates the development of the acetabulum. Average indicators of femoral head center position (AHI, LDA), achieved with surgery did not change significantly in the course of follow-up observations (see Table 2).

In group 1, after the removal of the external fixation, there were mostly joints with severe incongruence of the articular surfaces (Type IV). In five cases, in young children there was improvement of congruency during dynamic follow-p, which indicates a gradual adaptation of articular surfaces. In group 2, there were only three cases with severe incongruence of articular surfaces after removal of external fixation. The other joints corresponded to Type III. Improvement of congruence during following period was present in three cases (see Table 3). The distribution of joints by the types of congruence in group 1 was as follows: Type I, 2 joints; Type III, 4 joints, and Type IV, 3 joints. The distribution of types of congruence in group 2 was as follows: Type I, 1 joint; Type III, 6 joints, and Type IV, 1 joint.

The distribution of radiographic results in group 1 (including two reoperations) was as follows: good, 4; satisfactory, 4; and unsatisfactory, 1. The distribution in group 2 was as follows: good result — 3, satisfactory — 3, and unsatisfactory — 1 result.

Case study 1. Patient U., 9 years of age, female. The patient was admitted for pathological dislocation (Type IVa joint). Multistage surgical reconstruction of the hip joint was performed, which included transtrochanteric osteotomy, closed reduction of the femoral head in the acetabulum and iliac osteotomy. The follow-up examination 3 years after surgery showed that good articular relations in the hip joint had been maintained (Figure 1).

Case study 2. Patient K., 5 years of age, male. Admitted for treatment of pathological hip dislocation (Type IVb joint). The treatment consisted of double transtrochanteric osteotomy with open reduction of the femur and iliac osteotomy. Good anatomical and physiological results are maintained 6 years after treatment (Figure 2).


Almost 75% of complications from septic arthritis in children are hip lesions [5, 14]. Aggravating risk factors include age under 6 months, delay of treatment greater than 4 days, and proximal femur osteomyelitis. There are indirect and direct consequences of septic coxitis. Indirect consequences include growth plate lesions and aseptic necrosis of the femoral head, while direct consequences include cartilage lesions and destruction of the proximal femur. These changes are frequently accompanied by the disruption of articular relations [14].

Surgical treatment of destructive dislocations is recognized as problematic and questionable [7]. According to Betz et al., in children who had septic coxitis within the first three months of life, reconstructive surgery was found to exacerbate the condition of patients as it resulted in stiff, painful joints [15]. Accordingly, the authors recommended refraining from any surgical intervention, except for epiphysiodesis and limb lengthening [5, 14-17]. Although for this category of patients, pain is usually not severe and sufficient mobility of the joint is preserved over a long period of time, degenerative changes in the hip and knee joints, in sacral-lumbar spine region will progress, resulting in the development of poor posture [9]. These issues, as well as irreversible changes in the gluteal muscles and pronounced hypoplasia of the acetabulum, significantly complicate hip replacement surgery, which is likely the case of most of these patients [9].

There are a number of reconstructive techniques aimed at the formation of the hip joint. To form the weight-bearing surface of the femur, depending on the degree of destruction, surgeons use a stump of the femoral neck, the medial or top part of the greater trochanter, or the lesser trochanter [6, 18-20]. One of the main controversies is in determining the appropriate age limit to indicate these interventions [9]. Freeland et al. noted good and satisfactory results in reconstructive surgeries performed on patients aged 8 to 13 years [21]. Similar results were obtained by Belokrylov et al. [19]. Garkavenko and Pozdeev reported that 73% of school-aged and adolescent patients had positive long-term results for hip arthroplasty [22]. However, most authors agree that positive results are achievable only in young patients [3, 5, 7-9]. Choi et al. favor reconstructive surgery only in children younger than six years of age. In older patients, because of the risk of developing painful joint stiffness, corrective femur osteotomy is preferred [10]. Without minimizing the known advantages of this surgery, it should be noted, that if these procedures are performed before the age of 14 years, then the femur straightens in the process of growth at the place of the formed support angle, and will likely require reoperation [10, 23]. Similar changes are observed after trochanteric arthroplasty, which contribute to hip dislocation relapse [8]. Another known shortcoming of corrective femur osteotomy is the distortion of the anatomy of the proximal femur, which makes it difficult, and sometimes impossible, to perform the hip joint replacement [10]. In this paper, we present the technique of progressive hip extension, which should be considered similar to the well-known technique by Krumins et al., involving a double transtrochanteric osteotomy and repositioning of the intermediate fragment in the direction towards the bottom of the acetabulum [6, 24]. The main differentiating feature of our technique is that we only perform the osteotomy that involves forming a fragment, including the femoral neck stump, extra-articular formation of the proximal femur, and restoration of articular relations after correctional rotation of the fragments is complete.

One of the advantages of this technique is the simultaneous restoration and stabilization of the articular relations and compensation of femur shortening. The method of double transtrochanteric osteotomy that was used in 8 cases is based on the well-known surgical technique by Withman and Wagner [20, 25], but differs in that it uses an external fixation device, the Ilizarov apparatus. Analysis of the immediate and mid-term results did not show any significant age-related differences. There were no cases of joint stiffness or severe pain syndrome. This suggests the possibility of expanding the age limits for the use of such reconstructive procedures. In theory, the formation of congruent articular surfaces, stability of joint relations, may slow down the progression of degenerative changes. However, the relatively short period of observation and limited number of cases did not allow neither confirmation nor invalidation of our thesis. Therefore, this matter requires further study. The presented results give reason to assert that the use of reconstructive surgery in school-age children does not result in the decompensation and loss of the joint function. We agree with most other authors in that the reconstruction of articular components not only improves the function of the limb, but also makes hip joint replacement less complicated. Additionally, it is likely that the joint replacement surgery is significant for all patients with destructive hip dislocation [9, 10, 25]. The necessary condition for the use of this technique is that the surgeon has considerable experience with the use of external fixation devices.


Reconstructive surgery using the Ilizarov external fixation device in children with defects in the proximal femur allows for the correction of incongruent articular surfaces, without causing decompensation of the joint. These interventions may be considered as alternatives to palliative femur osteotomy in primary school-age children.

Funding information and conflict of interest

The authors have no conflicts of interest to declare. This work was supported by the Ilizarov Research Center for Restorative Traumatology and Orthopedics, Kurgan, Russian Federation.

About the authors

Mikhail P Teplenkiy

Russian Ilizarov Scientific Centre for Restorative Traumatology and Orthopaedics

Author for correspondence.

Russian Federation MD, PhD, professor. Highest category orthopaedic surgeon. Head of Joint Pathology Laboratory, Head of Department 11 (pediatric joint pathology). Russian Iizaov Scientific Centre.

Evgeny V Oleinikov

Russian Ilizarov Scientific Centre for Restorative Traumatology and Orthopaedics


Russian Federation MD, PhD, junior researcher of the Joint Pathology Laboratory, orthopaedic surgeon of the department 11 (pediatric joint pathology). Russian Iizaov Scientific Centre

Vyacheslav S Bunov

Russian Ilizarov Scientific Centre for Restorative Traumatology and Orthopaedics


Russian Federation MD, PhD, leading researcher of the Joint Pathology Laboratory. Russian Iizaov Scientific Centre.


  1. Miyahara HS, Helito CP, Oliva GB. Clinical and epidemiological characteristics of septic arthritis of the hip, 2006 to 2012, a seven-year review. Clinics. 2014;69(7):464-468. doi: 10.6061/clinics/2014(07)04.
  2. Choi IH, Pizzutillo PD, Bowen JR, et al. Sequelae and reconstruction after septic arthritis of the hip in infant. J Bone Jt Surg.1990;72-A(8):1150-1165.
  3. Hunka L, Said SE, MacKenzie DA, et al. Classification and Surgical Management of the Severe Sequelae of Septic Hips in Children. Clin Orthop. 1982;171:30-36. doi: 10.1097/00003086-198211000-00004.
  4. Крысь-Пугач А.П., Гук Ю.Н., Куценок Я.Б., и др. Наш опыт хирургического лечения ортопедических последствий гнойного коксита у детей // Травма. - 2008. - № 1. - Т. 9. - С. 66-70. [Kris-Pugach AP, Guk YN, Kucenok YB, et al. Nash opit hirurgicheskogo lecheniya ortopedicheskih posledstvii gnoinogo koksita u detei. Travma. 2008;1(9):66-70. (In Russ).]
  5. Forlin E, Milani C. Sequelae of Septic Arthritis of the Hip in Children. A New Classification and a Review of 41 Hips. J Pediatr Orthop. 2005;28(5):524-528. doi: 10.1097/bpo.0b013e31817bb079.
  6. Goel SC, Logani V. Management of Sequelae of septicarthritis of Hip in Infancy. JK-Practitioner. 2003;10(3):169-175.
  7. Baghdadi1 T, Saberi1 S, Eraghi1 AS, et al. Late Sequelae of Hip Septic Arthritis in Children. Acta Medica Iranica.2012;50(7):463-467.
  8. Abrishami S, Karami M, Karimi A, et al. Greater trochanteric preserving hip arthroplasty in the treatment of infantile septic arthritis: long-term results. J Child Orthop. 2010;4:137-141. doi: 10.1007/s11832-010-0238-x.
  9. Dobbs M, Sheridan JJ, Gordon JE, et al. Septic Arthritis of the Hip in infancy. Long-term follow-up. J Pediatr Orthop. 2003;23(2):162-168. doi: 10.1097/01241398-200303000-00006.
  10. Choi IH, Yoo WJ, Cho T-J, et al. Operative Reconstruction for Septic Arthritis of the Hip. Orthop Clin N Am. 2006;37:173-183. doi: 10.1016/j.ocl.2005.09.005.
  11. Colton CL. Chiari osteotomy for acetabular dysplasia J Bone Joint Surg. 1972-А;54-B(4):578-5898.
  12. Kruczynski J. Avascular necrosis of the proximal femur in developmental dislocation of the hip: Incidence, risk factors, sequelae and MR imaging for diagnosis and prognosis. Acta Orthop Scand. 1996(268);67:2-48. doi: 10.3109/17453679609155228.
  13. Mesa РА, Yamhure FH. Procedure DEGA type by double approach. Revista Colombiana de Ortopedia y Traumatología.2004;18(3):77-89.
  14. Sucato DJ, Schwend RM, Gillespie R. Septic Arthritis of the Hip in Children. J Am Acad Orthop Surg. 1997;5:249-260. doi: 10.5435/00124635-199709000-00003.
  15. Betz RR, Cooperman DR, Wopperer JM, et al. Late sequelae of septic arthritis of the hip in infancy and childhood. J Pediatr Orthop. 1990;10:365-372. doi: 10.1097/01241398-199005000-00014.
  16. Chang CH, Huang SC. Reconstruction for sequelae of septic hip in children. J Formos Med Assoe. 1997;96(5):353-358.
  17. Cheng JCY, Lam TP. Femoral lengthening after type IVB septic arthritis of the hip in children. J Pediatr Orthop. 1996;16:533-539. doi: 10.1097/01241398-199607000-00023.
  18. Ахтямов И.Ф., Абакаров А.А., Белецкий А.В., и др. Заболевания тазобедренного сустава у детей. - Казань, 2008. - 455 с. [Ahtyamov IF, Abakarov AA, Beleckii AV, et al. Zabolevaniya tazobedrennogo sustava u detei. Kazan; 2008;455 р. (In Russ).]
  19. Белокрылов Н.М., Гонина О.В., Полякова Н.В. Восстановление опороспособности при патологическом вывихе бедра в результате остеолиза его шейки и головки в детском возрасте // Травматология и ортопедия России. - 2007. - № 1. - Т. 43. - С. 63-67. [Belokrilov NM, Gonina OV, Polyakova NV. Vosstanovlenie oporosposobnosti pri patologicheskom vivihe bedra v rezulbtate osteoliza ego sheiki i golovki v detskom vozraste. Travmatologiya i ortopediya Rossii. 2007;1(43):63-67. (In Russ).]
  20. Соколовский А.М., Соколовский О.А. Патологический вывих бедра. - Минск: Высшая школа, 1997. - 208 с. [Sokolovskii AM, Sokolovskii OA. Patologicheskii vivih bedra. Minsk: Visshaya shkola; 1997:208. (In Russ).]
  21. Freeland AE, Sullivan DJ, Westin GW. Greater trochanteric hip arthroplasty in children with loss of the femoral head. J Bone Joint Surg (Am). 1980;62:1351-61.
  22. Гаркавенко Ю.Е., Поздеев А.П. Отдаленные функциональные результаты артропластики тазобедренного сустава у детей с последствиями острого гематогенного остеомиелита // Травматология и ортопедия России. - 2008. - Т. 50. - № 4. - С. 46-53. [Garkavenko YE, Pozdeev AP. Otdalennie funkcionalnie rezultati artroplastiki tazobedrennogo sustavau detei s posledstviyami ostrogo gematogennogo osteomielita. Travmatologiya i ortopediya Rossii. 2008;4(50):46-53. (In Russ).]
  23. Rozbruch R, Paley D, Bhave A, Herzenberg JE. Ilizarov Hip Reconstruction for the Late Sequelae of Infantile Hip Infection. J Bone Jt Surg Am. 2005;87(5):1007-1018. doi: 10.2106/JBJS.C.00713.
  24. Krumins M, Kalnins J, Lacis G. Reconstruction of proximal end of the femur after haemotogenous osteomyelitis. Pediatr Orthop. 1993;13:63-67. doi: 10.1097/01241398-199301000-00013.
  25. Benum P. Transposition of the apophysis of the greater trochanter for reconstruction of the femoral head after septic hip arthritis in children. Acta Orthopaedica. 2011;82(1):64-68. doi: 10.3109/17453674.2010.548030.
  26. Wagner H. Osteotomies for congenital hip dislocation. In The hip. Proceedings of the Fourth Open Sсientific Meeting of the Hip Society. St Louis, Mosby. 1976:43-61.



Abstract - 964

PDF (Russian) - 346

PDF (English) - 259




Copyright (c) 2016 Teplenkiy M.P., Oleinikov E.V., Bunov V.S.

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

This website uses cookies

You consent to our cookies if you continue to use our website.

About Cookies