Search tips
Search criteria 


Logo of jchildorthJournal of Children's Orthopaedics
J Child Orthop. 2009 June; 3(3): 223–227.
Published online 2009 May 5. doi:  10.1007/s11832-009-0170-0
PMCID: PMC2686811

Axial STIR MRI: a faster method for confirming femoral head reduction in DDH



Non-concentric reduction of the femoral head within the acetabulum is detrimental to its delicate cartilaginous structure and may result in a growth disturbance. Successful relocation of the dislocated femoral head depends on subtle clinical findings and radiography. The combination of a dislocated femoral head and a severely dysplastic acetabulum can result in a clinical examination that is unhelpful in confirming reduction under anaesthesia. In cases where uncertainty existed regarding head reduction in a spica cast, we performed axial short inversion time inversion recovery (STIR) and axial proton density magnetic resonance imaging (MRI) scans. We retrospectively reviewed the efficiency and accuracy of MRI in confirming femoral head location after closed reduction and spica application in eight children.


One hundred and fifty-three cases of developmental dysplasia were treated with examination under anaesthesia and spica application in our unit over a 3-year period. Eight cases where MRI scanning was performed were identified. Before application of the spica cast, we used radiographic screening to assess the stability of the reduction. Absence of the ossific nucleus within the femoral head made confirmation of the location with the image intensifier unreliable. To confirm concentric femoral head location after closed reduction and spica application, we performed an MRI scan in the immediate post-anaesthesia period.


All scans were performed within 30 min of application of the spica, and the average time for each scan was 5 min. All eight children who had MRI post-application of the spica had concentric reduction of the femoral head. MRI allowed three-dimensional appreciation of the acetabulum and femoral head. Use of the axial STIR images allows accurate assessment of the cartilaginous ossific nucleus. All patients were discharged the same afternoon and followed up as outpatients. No patient in our group required contrast arthrography.


While not indicated in all cases of femoral head dislocation, MRI is useful to confirm concentric reduction of the femoral head in a dysplastic acetabulum when examination under anaesthesia and radiographic screening have been uncertain. In our series, 1 in 20 cases needed MRI. This is a reliable, non-invasive method confirming definite reduction of the femoral head prior to discharge in all of our patients. In this initial series, all patients had axial and coronal STIR and proton density MRI. We now only use axial STIR images because they provide adequate information regarding the position of the femoral head relative to the acetabulum.

Keywords: Developmental hip dysplasia, Magnetic resonance imaging, Closed reduction


Non-concentric reduction of the femoral head within the acetabulum is detrimental to the delicate cartilaginous structure and may result in a growth disturbance. The goals of treatment in developmental dysplasia of the hip (DDH) are to obtain and maintain complete reduction and to avoid the development of a growth disturbance [1]. Assessment of the adequacy of the reduction after surgical management for DDH is technically difficult. The feel of the hip examination is key to its success and the movement of the hip in and out of its socket is a delicate event. The clinical examination is dependant on the ability of the surgeon to feel the femoral head reducing back into the acetabulum.

This clinical examination is complemented by a number of radiological investigations. Most common is the use of fluoroscopy and arthrography in theatre. We noted a subset of patients presenting with DDH who had an equivocal clinical examination or screening radiograph when in theatre. These children had a very dysplastic acetabulum and a dislocated or subluxed hip. In those children with an equivocal clinical examination, there was no feel for the femoral head dislocating or relocating, and there was no appreciable zone of stability. In the children with radiographs which suggested that the femoral head was not concentrically reduced, we had clinically appreciated a reduction of the femoral head into the acetabulum. The dilemma that faced us was what radiographic parameter would most reliably confirm femoral head reduction in the spica cast while avoiding general anaesthetic and with minimal radiation dose. In these children, we opted to perform an axial short inversion time inversion recovery (STIR) magnetic resonance imaging (MRI) scan without sedation. In this article, we highlight the advantages of MRI while addressing some of the limitations apparent in the literature.


Between 2003 and 2005, 153 children had spicas applied in our unit for a spectrum of developmental hip dysplasia, ranging from severe dysplasia to frank dislocation. Excluded from these figures are spicas applied post-Salters osteotomy and post-open reduction. One hundred and forty-five children had spica applied for dysplasia in the presence of a stable reduced femoral head or a definitely reducible femoral head. The femoral head was determined to be reducible when, from its dislocated position, it could be felt to reduce over the posterior rim of the acetabulum. Closed reduction was performed under general anaesthesia with longitudinal traction, flexion and abduction of the affected hip, while lifting the greater trochanter anteriorly. Our criteria for performing an MRI post-application of the spica were if the clinical examination did not confirm reduction of the femoral head into the acetabulum or if, after clinical reduction, the radiograph did not confirm the reduction and had the proximal femoral physis pointing away from the triradiate cartilage.

In our series over a 3-year period, eight children (mean age 19 weeks, range 12–31 weeks) had an equivocal clinical examination under anaesthesia or had a clinically reduced femoral head and radiographs that did not confirm reduction. In the children with an equivocal clinical examination, we were unable to feel the ‘clunk’ of a relocated femoral head into the acetabulum and were, therefore, unable to determine a safe zone of stability. Screening radiography was also performed in theatre pre- and post-application of the spica. The femoral head is considered to be reduced when the proximal femoral physis is opposite the triradiate cartilage. The decision to proceed to MRI is made based on the clinical and/or radiographic findings pre- and post-spica application. The hips are radiographically screened prior to reduction and again post-reduction, with the hip held in situ by the senior author. This image is saved and compared with the post-spica application image. In these eight cases, we felt that we could not safely rely on the screening radiograph to confirm the reduction. We did not perform arthrograms routinely in these patients. We felt that performing the arthrogram pre-application of the spica may be of benefit, but that this benefit would be lost post-application of the spica. Also, obtaining the necessary views to confirm concentric location of the femoral head can be difficult with the spica cast in situ [2].

The children were assessed by a consultant anaesthesist and deemed safe for transfer from the recovery unit direct to the MRI scanner. We used a 1.5-T Signa MRI scanner and a phased array surface coil. We performed axial and coronal STIR and proton density images on all eight children. We used a fast axial STIR sequence with a repetition time of 4,150, an echo time of 50 and an inversion time of 150. The field of view was set at 320 mm. We chose axial STIR because it is a useful sequence in musculoskeletal imaging since fat is suppressed. For the axial STIR images, the scan plane was set off a coronal localiser and set parallel to the femoral heads. The slices were 3-mm slices with no gap. A total of 16 slices covered the entire acetabulum.


All MRI scans were reviewed by the senior author and a consultant radiologist. Femoral head reduction was confirmed if the capital femoral epiphyses was in contact with the posterior margin of the acetabulum, even if it is asymmetric with the other side [3]. All eight patients referred for MRI had successful scans. All examinations were diagnostic, even though two were moderately degraded by motion artefact. No scan had to be repeated. The time taken to complete each MRI scan was less than 5 min. All femoral heads were reduced and centrally located in the acetabulum. The MRI gave excellent three-dimensional views of the acetabulum, allowing careful documentation of the dysplasia for repeated spica applications. We were able to appreciate posterior and anterior dysplasia that can be overlooked on the image intensifier, where our screening images are two-dimensional (Figs. 13).

Fig. 2
Her position in the spica cast after closed reduction. We found it difficult to determine whether the femoral heads are reduced because the physis is not pointing at the triradiate cartilage
Fig. 1
Screening radiograph of a child with bilateral dysplasia and subluxed hips
Fig. 3
An axial short inversion time inversion recovery (STIR) magnetic resonance imaging (MRI) image performed directly after recovery confirming bilateral femoral head reduction in the same child

No child needed sedation for scan completion. While two scans were moderately degraded by movement artefact, confirmation of femoral head reduction was still possible.


There are many radiological modalities available to assess the position of the femoral head in the acetabulum. Plain radiographs give a two-dimensional appreciation of the capital femoral epiphyses in relation to the acetabulum [2]. When the femoral epiphyses has not ossified, this interpretation is even more difficult and the relationship between the femoral head and the acetabulum can only be extrapolated. Even when the ossific nucleus is present, it cannot be assumed that it is in the centre of the capital femoral epiphyses.

Ultrasound is a safe modality which has shown that real-time multi-planar imaging is possible. However, it is less helpful in children in whom the ossify nucleus has appeared because acoustic shadows obscure a large part of the acetabulum [4]. Computed tomography involves a large dose of radiation; 0.11–0.5 rad [5]. It has a limited ability to identify the soft tissues and the non-ossified femoral head. It does, however, give reliable three-dimensional information regarding the relative positions of the proximal femur and bony acetabulum.

Arthrography is the yardstick against which all other techniques are assessed [6]. It is an established method to guarantee successful reduction and rule out obstruction that may prevent deep concentric reduction. It indirectly visualises structures such as ligamentum teres and transverse acetabular ligament that can impede reduction. Arthrography is widely used before reduction, but it cannot be repeated once the spica is applied [2]. Also, it is unwise to assume that the position seen on arthrography equates to that in the spica. The question with arthrography is how many views are necessary to evaluate reduction. Crawford and Carothers recommended eight views, while Lonnerholm and Ozonoff both recommend six. Drummond proposed three views; neutral, von Rosen and frog lateral. The issue is that these views cannot be repeated in children in spica casts [2].

MRI has been shown to accurately depict normal anatomy of the infant hip. It has also been shown to be an accurate tool for assessing the adequacy of reduction on DDH. MRI can be used to evaluate the result of closed or open reduction in patients when difficulty arises after the application of a spica [7]. It provides excellent definition of the cartilaginous acetabulum and femoral head and allows detailed assessment of their relationship to one another [8]. It provides information in more than one plane when compared with conventional arthrography.

The technique of MRI has a number of reported limitations, most notably the time taken to perform the examination. This ranges in the literature from 3 to 45 min. Laor et al. reported times between 2 min 30 s to 6 min for completion of the scan. Jaramillo et al. averaged 15 min, while Bos et al. and Guidera et al. averaged 25 and 45 min, respectively. All of the scans were completed in less than 5 min in this study. It is also important to note that sedation had been necessary in a number of studies [3, 911]. We did not use sedation in any of the children in the MRI scanner. Cost has also been a cause for some concern, but the advantage of a confirmed concentric reduction in plaster cast justifies the additional expense because serious pitfalls can be avoided [12]. There are a number of image sequences available and reported in the literature regarding MRI and DDH. Studies have reported their experience with coronal and axial images, different field strengths, body coils versus surface coils and different image sequences [2, 3, 8, 13]. We used a 1.5-T field strength machine, which gives better resolution than 0.5 T. Also, a surface coil improves soft-tissue definition compared with a body coil. We also found that axial STIR images were sufficient to confirm femoral head reduction. We no longer perform coronal STIR or proton density images.

In our unit, we perform axial STIR MRI when we cannot be sure that we have obtained concentric femoral head reduction based on our clinical examination and screening radiography. Inability to clinically appreciate the reduction of a dislocated femoral head is the exception rather than the rule. We feel that it is not necessary to subject all children post-closed reduction of a femoral head to MRI and it would also be an unnecessary use of resources.


While not indicated in all cases of examination under anaesthesia and spica application in developmental dysplasia, there is a role for magnetic resonance imaging (MRI) in confirmation of the location of the femoral head in a subset of affected children post-closed reduction and application of the spica. In our series, 5.2% of all new patients with developmental dysplasia of the hip benefited from the availability of an MRI scanner. This was facilitated by completion of the MRI scan in the immediate post-anaesthesia period and movement artefact was reduced by the presence of the spica. In our series, all patients had adequate scans without sedation. In this series of eight patients, each femoral head was concentrically reduced in the spica cast and this was demonstrated on axial short inversion time inversion recovery (STIR) MRI. We intend to continue to use MRI for the confirmation of femoral head reduction in the occasional case where concentric reduction of the femoral head is in doubt. The time taken from entry to the scanner to completion of the scan was less than 5 min. We use MRI because it is a fast, non-invasive procedure, with high sensitivity and specificity, with no ionising radiation and it provides a three-dimensional view of the femoral head and acetabulum.

Axial STIR MRI is a useful modality for confirming concentric femoral head reduction in selected cases of developmental dysplasia of the hip (DDH).


I would like to thank R. O. Connor and all the staff in Department of Paediatric Radiology, Temple Street, Dublin for helping with the preparation of this article.


1. Weinstein SL. Developmental hip dysplasia and dislocation. In: Morrissey RT, Weinstein SL, editors. Pediatric orthopaedics. 4. Philadelphia: Lippincott-Raven; 1996. pp. 903–950.
2. Duffy CM, Taylor FN, Coleman L, Graham HK, Nattrass GR. Magnetic resonance imaging evaluation of surgical management in developmental dysplasia of the hip in childhood. J Pediatr Orthop. 2002;22:92–100. [PubMed]
3. Laor T, Roy DR, Mehlman CT. Limited magnetic resonance imaging examination after surgical reduction of developmental dysplasia of the hip. J Pediatr Orthop. 2000;20:572–574. doi: 10.1097/01241398-200009000-00005. [PubMed] [Cross Ref]
4. Johnson ND, Wood BP, Noh KS, Jackman KV, Westesson PL, Katzberg RW. MR imaging anatomy of the infant hip. AJR Am J Roentgenol. 1989;153:127–133. doi: 10.2214/ajr.153.1.127. [PubMed] [Cross Ref]
5. Smith BG, Millis MB, Hey LA, Jaramillo D, Kasser JR. Postreduction computed tomography in developmental dislocation of the hip: part II: predictive value for outcome. J Pediatr Orthop. 1997;17:631–636. doi: 10.1097/01241398-199709000-00011. [PubMed] [Cross Ref]
6. Westhoff B, Wild A, Seller K, Krauspe R. Magnetic resonance imaging after reduction for congenital dislocation of the hip. Arch Orthop Trauma Surg. 2003;123:289–292. doi: 10.1007/s00402-003-0518-8. [PubMed] [Cross Ref]
7. Fisher R, O’Brien TS, Davis KM. Magnetic resonance imaging in congenital dysplasia of the hip. J Pediatr Orthop. 1991;11:617–622. doi: 10.1097/01241398-199109000-00011. [PubMed] [Cross Ref]
8. McNally EG, Tasker A, Benson MK. MRI after operative reduction for developmental dysplasia of the hip. J Bone Joint Surg Br. 1997;79:724–726. doi: 10.1302/0301-620X.79B5.7772. [PubMed] [Cross Ref]
9. Aoki K, Mitani S, Asaumi K, Akazawa H, Inoue H. Utility of MRI in detecting obstacles to reduction in developmental dysplasia of the hip: comparison with two-directional arthrography and correlation with intraoperative findings. Orthop Sci. 1999;4:255–263. doi: 10.1007/s007760050101. [PubMed] [Cross Ref]
10. Bos CFA, Bloem JL, Obermann WR, Rozing PM. Magnetic resonance imaging in congenital dislocation of the hip. J Bone Joint Surg Br. 1988;70:174–178. [PubMed]
11. Guidera KJ, Einbecker ME, Berman CG, Ogden JA, Arrington JA, Murtagh R. Magnetic resonance imaging evaluation of congenital dislocation of the hips. Clin Orthop Relat Res. 1990;261:96–101. [PubMed]
12. Tennant S, Kinmont C, Lamb G, Gedroyc W, Hunt DM. The use of dynamic interventional MRI in developmental dysplasia of the hip. J Bone Joint Surg Br. 1999;81:392–397. doi: 10.1302/0301-620X.81B3.8964. [PubMed] [Cross Ref]
13. Jaramillo D, Villegas-Medina O, Laor T, Shapiro F, Millis MB. Gadolinium-enhanced MR imaging of pediatric patients after reduction of dysplastic hips: assessment of femoral head position, factors impeding reduction, and femoral head ischemia. AJR Am J Roentgenol. 1998;170:1633–1637. doi: 10.2214/ajr.170.6.9609187. [PubMed] [Cross Ref]

Articles from Journal of Children's Orthopaedics are provided here courtesy of EPOS