September 11, 2017
16 min read

Surgical hip dislocation in children is safe for muscles, preserves blood supply

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

This Orthopedics Today Round Table features surgical hip dislocation in children, a procedure described by Reinhold Ganz, MD, about 17 years ago. Surgical hip dislocation allows an unrestricted 360° view of the hip through direct access, spares the surrounding muscles and preserves the blood supply.

Conditions treated with the surgical hip dislocation (SHD) technique include, but are not limited to, femoro-acetabular impingement (FAI), slipped capital femoral epiphysis (SCFE), Legg-Calvé-Perthes (LCP), tumors, trauma and other congenital or acquired conditions. Our participants, all of whom are pediatric orthopedic surgeons who have considerable experience with SHD, present their individual preoperative strategies. They discuss indications, individual workup and why particular imaging modalities are needed and what each provides. They also point out the advantage of the procedure when patients and their families inquire about less-invasive or minimally invasive surgery (MIS) and arthroscopic approaches.

Among other areas covered are the consenting process and drawings of the anatomy for the family; the role of preoperative rehabilitation, if any; what to expect in terms of recovery; postoperative hip range of motion (ROM) and weight-bearing; the timing of return to usual activities and sports; and potential complications.

The SHD technique has not been without occasionally reported complications. The panel presents opinions regarding the need for intraoperative femoral head vascular monitoring and techniques to prevent trochanteric nonunions or nerve injuries.

We hope the information conveyed herein helps Orthopedics Today readers learn about SHD and inspires them to use it in appropriate circumstances.

Alvin H. Crawford, MD, Hon Causa Gr, FACS

Alvin H. Crawford, MD, Hon Causa Gr, FACS: What was your motivation to first perform the SHD approach or did you pursue this approach during or since your fellowship training? Is it used for all conditions that require an intra-capsular approach?

Roundtable Participants

  • Alvin H. Crawford
  • Moderator

  • Alvin H. Crawford, MD, Hon Causa Gr, FACS
  • Cincinnati
  • Eduardo Novais
  • Eduardo N. Novais, MD
  • Boston
  • Wudbhav Sankar
  • Wudbhav N. Sankar, MD
  • Philadelphia
  • Patrick Whitlock
  • Patrick W. Whitlock, MD, PhD
  • Cincinnati
  • Ira Zaltz
  • Ira Zaltz, MD
  • Detroit

Eduardo N. Novais, MD: My motivation to perform the surgical dislocation approach was based on the versatility of this approach, which allows full exposure of the proximal femur, as well as the acetabulum. During my fellowship at the Mayo Clinic with Franklin H. Sim, MD, I learned that the surgical dislocation approach is helpful in some clinical scenarios beyond FAI, including benign tumors of the proximal femur, such as chondroblastoma of the femoral head and osteochondroma of the femoral head and neck; synovial chondromatosis of the hip; and pigmented villonodular synovitis (PVNS). However, the SHD was initially described for the treatment of FAI. In fact, the concept of FAI developed after Ganz developed this surgical approach. During my fellowship with Michael B. Millis, MD, and Young-Jo Kim, MD, PhD, I was exposed to the versatility of SHD. Currently, I use this approach for the treatment of complex FAI deformity, healed deformities associated with LCP and SCFE. I also use it in multiple trauma situations in the hip and pelvis, including for fractures of the femoral head, traumatic dislocation of the hip and fractures of the posterior acetabular wall.


When I choose to perform a modified Dunn procedure, I spend a good amount of time with the family explaining to them this procedure is not the usual standard of care and may result in some disability related to inherent complications associated with SCFE treatment. However, when such cases are performed in highly specialized centers, the technique has a relatively low risk of complications. When there is a specific indication to perform an operation and I choose to use the SHD approach and the family has questions about MIS, I explain that this approach allows for better exposure of the hip. This question is mainly raised when treating FAI. Although most studies have not found major differences in outcomes between a SHD approach and an arthroscopic approach for the treatment of FAI impingement, hip arthroscopy has gained popularity. In my practice, SHD helps with access to the superior and postero-superior aspect of the femoral head-neck junction. Therefore, when I see the patient with a major cam deformity that requires me to trim off the cam deformity and close to the retinacular vessels, I prefer the SHD approach. However, for regular FAI cam deformities, an arthroscopic approach is becoming more popular because the recovery is easier in the short term.

Ira Zaltz, MD: I was introduced to the periacetabular osteotomy (PAO) and the concept of acetabular reorientation for residual acetabular dysplasia by Michael B. Millis, MD, when I was a resident, at which time I developed an interest in hip surgery. During my fellowship, hip preservation surgery was not a defined area of surgical interest. None of the faculty where I was a fellow performed the PAO, and since we only treated patients who were younger than 18 years, there was not much interest in or appreciation for young adults with residual pediatric disease. Furthermore, the concept of impingement had not been described. By happenstance, when I entered practice in Detroit, Jeffrey W. Mast, MD, was relocating and left both a large local awareness among orthopedic surgeons that young adults with hip deformity were treatable, and a large group of patients. Therefore, I had the opportunity to get involved in the field from the outset of my career. Over time, I delved deeply into the literature and attended the AO hip meetings where Ganz introduced the concept of FAI and the SHD technique. I watched Ganz’ live surgical videos at several AO courses, practiced the procedure on cadaveric specimens and gradually began to use the procedure to treat FAI. Over time, I began to incorporate the procedure into most pediatric osteotomies used for SCFE, LPC, post-avascular necrosis (AVN), post-developmental dysplastic hip (DDH), upper femoral deformity, femoral chondral defects, and, FAI. Currently, I use arthroscopy for about two-thirds of my FAI cases; however, I still use the dislocation approach for patients with large femoral deformities, complex cases of FAI and deformities associated with the spectrum of pediatric diseases, including SCFE, LPC and DDH.


Patrick W. Whitlock, MD, PhD: I was introduced to the approach as a resident. My first exposure to the approach and to its associated procedures was for the treatment of LCP disease, as a means to perform a proximal femoral osteoplasty (PFO), an extended retinacular flap with relative femoral neck lengthening and labral repair (to address coxa magna, coxa breva, intra-articular/extra-articular impingement, labral pathology) in conjunction with the PAO for residual post LCP disease dysplasia. Thereafter, I assisted in the approach and development of the associated extended retinacular flap as part of the modified Dunn procedure, in addition to its use in addressing a large cam deformity and some pincer FAI in the absence of significant retroversion. This exposure and related experience continued throughout my pediatric and hip preservation fellowships.

I think an anterior or lateral approach to the hip is useful for other procedures, such as septic arthritis and PFO. I also use arthroscopy to address FAI or other intra-articular pathologies when I am confident I can address the pathology sufficiently without the need for SHD. In this respect, I typically reserve the SHD approach for intra-articular pathology that I am unable to confidently approach arthroscopically (diffuse PVNS, large circumferential cams, femoral head defects/fractures and associated posterior acetabular wall fractures), in addition to the modified Dunn procedure.

Figure 1. This 15-year-old male patient with a history of DDH underwent closed reduction as an infant, but presented with increasing pain and abductor dysfunction.
Figure 2. This radiograph shows the results 18 months after SHD, femoral head-neck osteoplasty and a relative neck lengthening.

Images: Sankar WN

Wudbhav N. Sankar, MD: The surgical dislocation is an essential tool in the toolbox for anyone treating complex hip disorders in adolescents and young adults. It allows wide access to the hip and enables a host of different procedures to be performed, including near-circumferential osteoplasty for large cam lesions, rim trimming, capital realignment for SCFE, relative neck lengthenings and femoral head reduction, among others. I learned this procedure during my fellowship and use it frequently in my practice. Although it is a powerful tool, I do not use it for all conditions requiring an intra-capsular approach. However, I find it complementary to direct anterior approaches and surgical arthroscopy of the hip. The mechanical implications of various pediatric hip deformities have been better understood through our experience with the SHD because this approach allows the surgeon to directly observe the dynamic relationship of the proximal femur with the acetabulum intraoperatively. For these reasons, in complex situations I typically advise families that this more utilitarian approach, rather than a minimally invasive one, is often needed for certain pathologies that require careful intraoperative assessment and on-the-spot decision-making. My preoperative workup and need for advanced imaging depends on the underlying disease process rather than the surgical approach. The imaging findings often influence my decision to choose the SHD approach.


Crawford: Has using this approach changed your preoperative workup, such as tertiary imaging, or the consent language used with the family?

Whitlock: The associated pathology is often complex and requires MRI and 3-D CT in my hands. I emphasize there is a risk of AVN with the approach, specifically with the development of the extended retinacular flap and epiphyseal reorientation. However, I also explain that, in the absence of these two procedures, there is minimal risk of AVN.

Zaltz: The procedure itself has not changed my preoperative workup; however, gaining experience treating FAI has radically changed our thought process and influenced our workup. At one point, early in the FAI experience, it was thought hips with a low ROM were impinging hips. However, we now understand that even a hip with high-grade dysplastic acetabuli can have a low ROM and, alone, motion is insufficient for the diagnosis of FAI or instability. Therefore, we have become more reliant upon 3-D imaging to carefully analyze the coverage of the femoral head by the acetabulum, as well as the upper femoral morphology. We also understand morphologic features frequently coexist with instability and FAI in the same patient.

Figure 3. The radiographs are of a 12-year-old female patient with SCFE.
Figure 4. The patient was treated with a modified Dunn procedure and was asymptomatic at the 2-year follow-up.

Images: Zaltz I

My consent has not changed. I explain the procedure carefully to patients and use a model to demonstrate the technique. We have shown the procedure is safe with a complication rate comparable to arthroscopy. Furthermore, the procedure is not terribly morbid when performed properly and patients usually experience a rapid recovery.

Crawford: How do you respond to a family’s inquiry about a MIS procedure?

Whitlock: It is an option, specifically arthroscopy. However, in general, I explain my chances of a successful outcome and meeting expectations are greatest using this open approach in the setting of complex pathology. Usually patients and their families are satisfied with this explanation. Despite this caution, I have had patients with pathology I believed could not be addressed arthroscopically by me pursue treatment elsewhere strictly for the MIS approach, the results of which may be variable.

Zaltz: As I treat adult and pediatric patients, I am frequently asked this question. Based on data from our research and outcomes from Bryan T. Kelly, MD, at Hospital for Special Surgery, I believe certain patients with FAI are more effectively treated with an open approach. These include patients with large cam deformities that extend to the posterior femoral head; patients with low anteversion or any retroversion; patients with significant femoral chondral defects; and patients with coxa profunda. Of course, most pediatric-related deformities are not amenable to a less-invasive approach because either extensive reshaping or an osteotomy are necessary. Interestingly, the recovery following either arthroscopy or SHD are comparable, and I encounter patients who request an open approach because the arthroscopic technique has a bad reputation in certain on-line patient forums.


Crawford: What intraoperative perfusion monitoring technique do you use? What would prompt you to abort or change the procedure based on your monitoring response?

Novais: I use monitoring of the femoral head by the presence of active pulsatile bleeding and I also use Camino intracranial pressure (Integra) monitoring. We published on the accuracy of monitoring femoral head bleeding with these two techniques in Clinical Orthopaedics and Related Research in 2016. We found these techniques are not perfect but are helpful in in assessing blood flow to the femoral head during the SHD approach. I typically use monitoring for treatment of SCFE, treatment of healed LCP disease deformities and for treatment of traumatic posterior hip dislocation.

Figure 5. A 15-year-old male patient who had several months of right hip pain was referred with an MRI done with and without contrast. This coronal image shows diffuse PVNS of the right hip without intraosseous involvement and a large cam lesion.
Figure 6. Appearance on plain films of the hip after fixation of the trochanteric osteotomy. After complete SHD, a complete synovectomy was done except in the area of the retinacular vessels and posterior superiorly-inferiorly near the short external rotators and the origin of the medial femoral circumflex artery.
Figure 7. An anteroposterior (AP) pelvis radiograph at 15 months postoperatively shows the joint space is maintained. The greater trochanteric osteotomy is healed and at that time, there was no evidence of a hardware complication. The patient is asymptomatic and able to perform all desired activities.

Images: Whitlock PW

Zaltz: I do not routinely use intraoperative perfusion monitoring techniques. When I perform a modified Dunn procedure for displaced, unstable SCFE, I typically inspect the retinacular insertion and will then occasionally make a 0.62-cm drill hole in the anterolateral epiphysis (the so-called “watershed” zone) to observe for bleeding. The issue with monitoring is it is not completely reliable, but it can be immediately reassuring for the surgeon.

Sankar: I only monitor intraoperative perfusion in cases where I am developing a full extended retinacular flap (such as a modified Dunn or head reduction). In these cases, I would not abort a case based on monitoring, but would use the findings to adjust parts of the case that are modifiable, such as the tension on a head reduction or looseness of the capsular closure.

Whitlock: I use Doppler, but I do not think it is a question of aborting the procedure in reference to the modified Dunn, but reorienting the position of the epiphysis, or shortening the femoral neck to accommodate any remodeling of the retinacular flap and vessels which has occurred.


Crawford: What is your postoperative rehabilitation to return to full, unrestricted weight-bearing and/or return to activities and athletics? Do you allow contact-impact activities after a certain amount of time or is that prohibited?

Sankar: For most surgical dislocations, I keep patients toe-touch weight-bearing (TTWB) for 4 weeks and then weight-bearing as tolerated with crutches for 4 additional weeks to protect the abductors. If a step cut is used for the trochanteric osteotomy, then this timeline can be accelerated by 2 weeks to 4 weeks. Return to athletics depends on bony healing and return of muscular strength and coordination, which is typically at 5 months to 6 months in my practice.

Novais: The postoperative rehabilitation program depends upon the type of pathology treated. For instance, if I am treating a FAI cam-type deformity, typically where SHD was performed using the step-cut osteotomy of the trochanter to allow for a more rapid healing of the trochanter, I start rehabilitation sooner. In this scenario, the patient will typically use crutches for about 6 weeks and physical therapy (PT) will start 2 weeks following surgery. The PT will be in three phases. Typically, the patient can return to sports at about 8 months after surgery. We published in the Journal of the American Academy of Orthopaedic Surgeons in 2016 a series of young athletes who underwent FAI treatment using the SHD approach. They had a return to play of about 85% with a mean return to play at about 9 months postoperatively. However, when treating other conditions, such as an unstable SCFE, I keep the patient non-weight-bearing for a minimum of 8 weeks followed by protected weight-bearing for a minimum of 4 weeks to 6 weeks. In surgical situations where there is a risk of dislocation, including when I repair a posterior wall fracture using the posterior hip dislocation approach in a child or adolescent patient, I would keep the patient in a brace for the first 6 weeks after surgery.

Whitlock: I am cautious. Up to 6 weeks postoperatively, patients are allowed touch-down weight-bearing (90° hip flexion, 10° in remainder of hip motion). At 6 weeks, they then progress to weight-bearing as tolerated and ROM as tolerated with supervised PT for the next 6 weeks and then begin strengthening at 3 months, which is followed by return to activity on an individualized basis, depending on the etiology. I caution patients with high ROM/repetitive activity/contact activity/prolonged impingement positions that their hip may not be the best for these types of activities, but pursuing those activities remains a personal and philosophical choice. I follow these patients on a long-term basis and expect to evaluate them throughout their lifetime and share in decision-making as they age and their lives and activities change.


Zaltz: When I perform a SHD for treatment of complex FAI, I usually perform a step-cut osteotomy of the greater trochanter that will allow the patient to bear weight as tolerated using crutches to protect the osteotomy and by limiting the abductor moment. When I perform a dislocation procedure that includes either an osteotomy; extensive cartilage restoration, such as osteochondral autograft transplantation system, autologous matrix-induced chondrogenesis and microfracture; and/or extensive labral repair, I recommend TTWB for 6 weeks. Then, after healing, I permit transition to full weight-bearing over 1 week to 2 weeks. Following a modified Dunn procedure, I usually have the patient use a wheelchair for 6 weeks, and then allow TTWB for another 6 weeks. Once the patient can fully bear weight, I prescribe PT for strengthening and gait training. We use a knee CPM machine for hip patients regardless of the procedure, to limit hip muscle stiffness following surgery. I allow most patients to begin athletic-specific training when they have regained satisfactory abductor strength, usually around 3 months postoperatively.

Crawford: What techniques are appropriate to approach or prevent such complications of trochanteric nonunion as nerve injury, heterotopic bone formation, avascular insult and implant displacement, which are associated with SHD?

Zaltz: Following a properly executed surgical dislocation, complications are rare. In our multicenter series by E.L. Sink and colleagues in 2011 in the Journal of Bone and Joint Surgery, the incidence of such complications was 2.7%. The surgical technique varies substantially from surgeon to surgeon and, if executed properly, such complications are rare.

The step-cut osteotomy is stable and has a broad surface area. Since I started using this technique, I have never encountered a trochanteric nonunion. Prior to the step-cut technique, the rate of nonunion at our institution was 0.33% and all nonunions were successfully treated by open reduction and internal fixation. We identified patient factors in the nonunions we treated, including activity non-compliance, smoking and hyperparathyroidism. We have never had a nerve injury at our institution. The rate of nerve injury following arthroscopy is higher.

Figure 8. While playing basketball, a 14-year-old boy had a posterior hip dislocation, which is shown on an AP pelvic radiograph taken in the ER.
Figure 9. When the hip was reduced, it revealed a displaced femoral head fracture.
Figure 10. The femoral head and the extent of the femoral head fracture can be seen during SHD.
Figure 11. The pattern of a posterior lateral-periosteal sleeve avulsion can be seen, which is typical of young patients with incomplete ossification of the posterior acetabular wall.
Figure 12. The anatomic reduction of the femoral head fracture is shown in an AP radiograph. The trochanteric fixation shows no signs of femoral head necrosis.

Images: Novais EN


Heterotopic ossification (HO) occurs about 5% of the time following the SHD, but the majority are little flecks of bone that are Brooker type 1 located at the tip of the trochanter, and are clinically insignificant. It is our routine to administer 1 week of indomethacin (weight-dependent dose) to all patients with open hip surgery to minimize the risk of HO. We prophylax for 2 weeks following arthroscopy.

We have never had cases of AVN following SHD, except after the modified Dunn procedure for the treatment of SCFE. Implant breakage occurs in about 66% of patients with trochanteric nonunion. We have had patients break their fixation following the modified Dunn because they were weight-bearing prior to radiographically confirmed healing. All of these patients required reoperation.

Novais: The complication of trochanteric nonunion is rare in my experience. Although the procedure was initially described with a trochanteric, simple osteotomy and fixation with two 3.5-mm screws, I currently prefer to use a step-cut osteotomy of the trochanter and I do the fixation with two 4.5-mm screws applied under AO technique for compression. Using this technique, it is rare to see problems with the trochanter union. The incidence of nerve injury with this approach is rare, as well. One can easily recognize the location of and protect the sciatic nerve with this approach. Also, I do not have major problems with heterotopic bone formation, possibly because I use prophylaxis for HO formation. I typically have patients use naproxen twice a day for the first month after surgery. Aside from treatment of unstable SCFE, an osteonecrosis of the femoral head complication is rare using the SHD approach.

Whitlock: I have not had any significant cases of trochanteric nonunion. Overall, it is a rare complication. I perform the trochanteric osteotomy with a Ganz osteotome, which produces an irregular and interlocking fragment.

Nerve injury is also a rare complication that occurs in less than 1% of cases and it is one I have not specifically observed in relationship to the approach. However, it can be seen in combination with PAO, especially in complex hips, such as in a patient with a post-LCPD hip that has undergone multiple surgeries and is at risk with PAO. This risk is not specifically inherent to the surgical dislocation approach, but is a result of associated procedures and patient factors. HO has not been an issue, in my experience.

Implant displacement has not been an issue as it relates to the approach, but it mostly results from non-compliance with activity and weight-bearing restrictions in cases of unstable SCFE treated with the modified Dunn procedure, when the epiphysis loses fixation and revision is necessary. There is a supposed increased risk of AVN when that occurs, as expected, but often these cases remain unaffected. The outcome of such cases may be unpredictable.


Sankar: I typically use three 4.5-mm cannulated screws for trochanteric fixation and have only had a couple cases of trochanteric nonunion with this fixation and postoperative restrictions. Nerve injury is almost nonexistent in my experience. Heterotopic bone can be minimized by careful dissection of the gluteus minimus and the use of indomethacin postoperatively. For a standard surgical dislocation (ie, without a retinacular flap) the risk of AVN in a well-performed surgery is essentially zero. For more complicated cases requiring a flap, AVN can be minimized by meticulous technique and intraoperative monitoring. The risk of implant displacement depends on the pathology being treated. In my experience, this is usually limited to modified Dunn procedures as we showed in a previous study. I have been able to minimize implant complications in these types of surgeries by moving to two 6.5-mm cannulated screws for epiphyseal fixation.

Crawford: What insights do you have on the recent article by V.V. Upasani and colleagues in Clinical Orthopaedics and Related Research about instability as a complication after the modified Dunn procedure for SCFE?

Zaltz: This publication illustrates a rare and unique complication that can occur when the SHD approach is used to treat severe SCFE. This case series illustrated the capacity to induce instability of the hip joint following open reduction using the modified Dunn approach for severe, long-standing SCFE deformity. This instability is thought to relate to soft tissue laxity that results from shortening the neck of the femur to avoid excessive tension on the retinaculum during the reduction maneuver. In most cases, the instability was managed successfully by either casting or traction. Nevertheless, this can be a devastating complication.

Sankar: I agree. I would only add that, in this specific circumstance of a modified Dunn procedure, transferring the greater trochanter slightly distally and/or using a postoperative brace may be helpful to limit the risk for this devastating complication.

Disclosures: Crawford, Novais and Zaltz report no relevant financial disclosures. Sankar reports he receives royalties from Wolter Kluwer Health for edited textbooks. Whitlock reports he receives funding support from the Pediatric Orthopaedic Society of North America Angela S.M. Kuo Memorial Award.