Early Postoperative Transverse Pelvic Fracture: A New Complication Related to Revision Arthroplasty with an Uncemented Cup

Springer, Bryan D. MD; Berry, Daniel J. MD; Cabanela, Miguel E. MD; Hanssen, Arlen D. MD; Lewallen, David G. MD

Journal of Bone & Joint Surgery - American Volume:
doi: 10.2106/JBJS.E.00088
Scientific Articles

Background: Uncemented hemispherical cups are commonly used to revise failed acetabular total hip components, even in the presence of marked acetabular bone loss. The purpose of the present study was to report a new complication of acetabular component revision with an uncemented hemispherical cup.

Methods: We retrospectively reviewed the records of seven patients (seven hips) in whom an early postoperative transverse acetabular fracture had developed following the implantation of an uncemented trabecular metal cup for the revision of a failed acetabular component. All patients were female. The average age was 63.6 years. The reason for acetabular revision was aseptic loosening of the original cup in five patients and reimplantation after a previous resection arthroplasty in the remaining two. The average cup size used for revision was 58 mm. In two hips, additional modular acetabular metal augments were used to restore the acetabular rim.

Results: The average postoperative time to diagnosis of a transverse acetabular fracture was eight months. Five of the seven patients presented with a marked acute increase in pain and a new displaced transverse acetabular fracture (pelvic discontinuity) that was visible on plain radiographs. Two patients were asymptomatic but had a nondisplaced transverse acetabular fracture. In all seven patients, the trabecular metal socket appeared radiographically to be well fixed to part of the pelvis. The five patients with a displaced fracture were managed with additional surgery to stabilize the fracture.

Conclusions: To our knowledge, early postoperative transverse pelvic fractures following revision of the acetabular component have not been reported previously. The most likely causes of this complication are further weakening of the remaining pelvic bone stock as a result of the reaming required to obtain a secure fit of a large-diameter hemispherical socket and the cyclic stresses on the weakened bone with resumption of walking. It is unlikely that the fractures occurred intraoperatively because in each case the socket remained well fixed to one of the pelvic fragments.

Level of Evidence: Therapeutic Level IV. See Instructions to Authors for a complete description of levels of evidence.

Author Information

1 OrthoCarolina Hip and Knee Center, 1915 Randolph Road, Charlotte, NC 28209. E-mail address: bryan.springer@orthocarolina.com

2 Department of Orthopaedic Surgery, Mayo Clinic, 200 First Street S.W., Rochester, MN 55905

Article Outline

Revision total hip arthroplasty in the presence of marked acetabular bone loss remains a challenge for the reconstructive surgeon. As bone ingrowth surfaces have improved, it has become feasible to revise the acetabulum, even with marked bone loss, with a cementless hemispherical cup1.

The purpose of the present study was to report a new complication of revision total hip arthroplasty with an uncemented hemispherical socket: postoperative transverse acetabular fracture associated with a well-fixed acetabular component.

Back to Top | Article Outline

Materials and Methods

From June 2000 to July 2003, 585 acetabular revisions were performed at our institution with the use of a hemispherical trabecular metal acetabular component. Seven patients (seven hips) who underwent revision of a failed acetabular component had development of a postoperative transverse acetabular fracture associated with a well-fixed socket following revision of the acetabular component. All seven patients were female. The average age of the patients at the time of the fracture was 63.6 years (range, thirty-nine to seventy-six years). Five fractures involved the right hip, and two involved the left hip. The initial diagnosis that had led to hip arthroplasty was developmental dysplasia in four patients and degenerative joint disease in three patients. The reason for acetabular revision was aseptic loosening of the acetabular component in five patients and reimplantation following previous resection arthroplasty for the treatment of deep periprosthetic infection in two patients. The criteria described by Berry were used to diagnose pelvic discontinuity2. These criteria included a visible transverse pelvic fracture on anteroposterior pelvic or Judet radiographs; medial offset of the inferior part of the pelvis in relation to the superior part of the pelvis as seen by a break in the Kohler line; and rotation of the hemipelvis as indicated by asymmetry of the obturator ring on the true anteroposterior pelvic radiograph.

Details on the operation were retrieved from the patient records retrospectively (Table I). Each acetabular defect was categorized according to the classification system of the American Academy of Orthopaedic Surgeons (AAOS)3. There were three type-I (segmental) defects and four type-III (combined segmental and cavitary) defects. At the time of the initial revision, all patients were managed with a modular tantalum trabecular metal acetabular component (Zimmer, Warsaw, Indiana). The average outside diameter of the component used was 58 mm (range, 52 to 70 mm). The trabecular metal revision shell has multiple screw-holes, and extra holes can be drilled through the trabecular metal for additional screw fixation. The outer periphery of the shell is 2 mm wider than a true hemisphere to allow for a peripheral rim press-fit. In six of the seven patients, the acetabulum was reamed to the same size as the implanted component, thus allowing for a 2-mm press-fit at the periphery of the cup. One acetabulum was underreamed by 2 mm, allowing for a 4-mm press-fit at the periphery. Allograft particulate cancellous bone graft was used to fill residual contained acetabular bone defects in five of the seven patients. The graft was hand-packed into the defects. In four patients, additional demineralized bone matrix putty was added to the allograft bone and was placed into the contained defects. The amount of bone graft used in each hip is listed in Table I. The cups were impacted into place with a mallet with use of multiple blows of moderate force on an inserter supplied by the manufacturer. When necessary, rim impactors, also provided by the manufacturer, were used to fully seat the acetabular component against the floor of the acetabulum. Each acetabular shell was further secured with four or five screws that were placed into the acetabular dome and the posterior column (Table I). The liner was cemented into the shell in all patients because this generation of the component could not be used with a modular uncemented insert. Three of the liners used were constrained sockets, and two had elevated lips.

Substantial superior segmental rim defects remained in two patients after reaming. Both patients were managed with trabecular metal augments and particulate bone graft. No structural allograft bone was used to treat these defects. The augments were fixed to the pelvis with additional screws (three screws in one patient and five in the other) and were cemented to the roof of the shell.

Back to Top | Article Outline


None of the seven patients had any evidence of an acetabular fracture at the time of the operation or on the immediate postoperative radiographs. Five patients presented with an acute increase in groin pain without antecedent trauma at an average of 8.0 months (range, two to twenty-seven months) postoperatively. All five had evidence of a new displaced transverse acetabular fracture on radiographs. The other two patients were asymptomatic and were noted to have new nondisplaced fractures at the time of routine follow-up at one and seventeen months postoperatively. The previous radiographs had been unremarkable. The diagnosis of a transverse pelvic fracture was made on the basis of standard postoperative anteroposterior radiographs of the pelvis and anteroposterior and lateral radiographs of the hips. When the complication was suspected, Judet radiographs were made for all patients. No computed tomography scans were made.

Back to Top | Article Outline

The five patients with a symptomatic displaced transverse acetabular fracture underwent reoperation. At the time of surgery, all five patients had a visible fracture line and gross motion between the superior and inferior portions of the pelvis. In four of the five patients, the trabecular metal acetabular component was well fixed to the superior portion of the pelvis (ilium) (Figs. 1-A and 1-B, 1-C and 1-D). In one patient, the acetabular component was well fixed to the inferior portion of the pelvis (ischium).

Four of the five displaced fractures were treated with a pelvic reconstruction plate. In each instance, the fracture was exposed through a posterior approach and a pelvic reconstruction plate was contoured along the posterior column of the acetabulum. After the fracture was reduced, the plate was fixed cephalad and caudad to the fracture with screws. Particulate cancellous allograft was then packed around the fracture site. No additional surgery was performed at the site of the acetabular component. In the remaining patient, a reconstruction cage was placed into the well-fixed cup to bridge the fracture. The iliac flange of the cage was removed with a burr, and the ischial flange of the cage was slotted into the ischium. No ischial screws were used to secure the flange, but three extra screw-holes were created in the cup and the screws were directed inferiorly up the base of the ischium to supplement the flange. The liner was then cemented into the cup-cage construct. At the time of the latest follow-up, four of the five patients had radiographic evidence of a healed pelvic fracture and a stable, well-fixed cup.

The two asymptomatic patients were managed nonoperatively. Each patient was managed with a course of protected weight-bearing until there was radiographic evidence of fracture-healing. At the time of the latest follow-up, both patients remained asymptomatic and had radiographic evidence of fracture-healing with a stable, well-fixed acetabular component.

Figure. No caption a...
Figure. No caption a...
Figure. No caption a...
Figure. No caption a...
Back to Top | Article Outline

Two patients who underwent stabilization of the fracture with a posterior column plate subsequently sustained a fracture of the fixation plate. One of these patients was managed with protected weight-bearing on crutches, and the fracture went on to heal despite the broken plate. At the time of the latest follow-up, at eighteen months, the patient reported no pain with weight-bearing activities. In the other patient, the plate fractured eight weeks postoperatively. At the time of the latest follow-up, the patient had an ongoing nonunion of the pelvic fracture. The acetabular component remained well fixed radiographically. At the time of the present report, additional surgery had been performed. No other patient experienced any additional postoperative complications related to the surgery.

Back to Top | Article Outline


The present report describes seven patients in whom a postoperative transverse acetabular fracture developed after revision hip arthroplasty with use of a hemispherical uncemented trabecular metal cup that remained well fixed to the pelvic bone. To our knowledge, this is the first report of this complication. Pelvic discontinuity can present in hips with an associated loose acetabular component and substantial acetabular bone loss2,4-11. Mahoney and Garvin reported a single case of post-revision pelvic discontinuity that developed in association with a pelvic stress fracture12. The patient in that report had been managed with revision of a loose cup, and the fracture was attributed to excessive intraoperative reaming of the acetabulum. In contrast, the fractures described in the present report appear to have occurred postoperatively. No patient had evidence of a pelvic fracture intraoperatively or on immediate postoperative radiographs, but all seven patients had subsequent radiographic evidence of a pelvic fracture at an average of 8.3 months postoperatively. Five of these patients presented with an acute onset of groin pain. At the time of reoperation, the surgeon confirmed a new transverse acetabular fracture. While it is not possible to completely exclude the possibility that some of these fractures occurred intraoperatively but were not manifested until months later, the time-course of presentation and the lack of clinical or radiographic signs immediately after the operation suggest that this is unlikely.

We hypothesize that with the resumption of activities of daily living, the cyclic stresses through bone that was deficient following the previous failed arthroplasty and further weakened by reaming for the new prosthesis led to the fracture. When the pelvis is loaded postoperatively, stress may be concentrated at the junction between bone that is rigidly fixed to the implant with screws (usually the superior portion of the pelvis) and bone that may undergo micromotion under load (usually the inferior portion of the pelvis). This proposed mechanism could explain both the occurrence of these postoperative fractures and the fact that the cup remained well fixed to part of the pelvis in all cases. While it is notable that all of these complications occurred in association with the use of a trabecular metal shell that was fixed with multiple screws, we do not believe that this complication is uniquely related to this implant design. Rather, we believe that the association of this complication with this design is related to our recent use of this component in difficult revision situations associated with severe bone loss and to the capability of this cup to gain fixation under these circumstances. This shell design, with a high-friction metal surface and the potential for augmentation of fixation with the placement of multiple screws through additional holes that the surgeon can create intraoperatively, has allowed us to achieve stable cup fixation, even in very severely deficient pelvic bone. The simplicity and success of the technique has led us to use it even when bone deficiency is severe and even when reaming of the anterior and posterior columns is necessary to gain good cup contact against host bone. However, this practice may leave the pelvic bone susceptible to fracture in certain patients.

In a previous report, Berry found female gender (p < 0.001) and rheumatoid arthritis (p = 0.003) to be significant risk factors in the development of pelvic discontinuity2.

The majority of patients in the present series had relatively small acetabula: all seven patients were female, and four of the seven had a dysplastic acetabulum secondary to developmental dysplasia. Thus, the cups used in our series were relatively large for the acetabula into which they were implanted. To ensure fixation of the hemispherical acetabular component, further reaming was required, which resulted in some loss of bone in all seven hips.

To reduce the likelihood of this complication, we recommend limiting the amount of columnar support that is removed during reaming of the acetabulum. In high-risk patients, limiting early postoperative weight-bearing also may reduce the risk of early fracture.▪

A commentary is available with the electronic versions of this article, on our web site () and on our quarterly CD-ROM (call our subscription department, at 781-449-9780, to order the CD-ROM).

In support of their research or preparation of this manuscript, one or more of the authors received grants or outside funding from Zimmer, Implex (A.D.H., D.G.L.), Stryker (M.E.C.), and DePuy (D.J.B.). In addition, one or more of the authors received payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity (Zimmer; Implex [A.D.H., D.G.L.]; Stryker [M.E.C.]; and DePuy [D.J.B.]). No commercial entity paid or directed, or agreed to pay or direct, any benefits to any research fund, foundation, educational institution, or other charitable or nonprofit organization with which the authors are affiliated or associated.

Investigation performed at the Department of Orthopaedic Surgery, Mayo Clinic, Rochester, Minnesota

1. , Berry DJ, Harmsen WS. Extra-large uncemented hemispherical acetabular components for revision total hip arthroplasty. J Bone Joint Surg Am. 2001;83: 1352-7.
2. . Identification and management of pelvic discontinuity. Orthopedics. 2001;24: 881-2.
3. . Periprosthetic bone loss of the acetabulum. Classification and management. Orthop Clin North Am. 1992;23: 279-90.
4. , Ali A. Periprosthetic fractures of the acetabulum. Instr Course Lect. 2004;53: 93-8.
5. , Muller C, Ganz R. Revision surgery in pelvic discontinuity: an analysis of seven patients. Clin Orthop Relat Res. 2002;398: 136-45.
6. , Coventry MB. Femoral and pelvic fractures after total hip arthroplasty. J Bone Joint Surg Am. 1974;56: 483-92.
7. , Hungerford DS. Stress fracture of the pubis after total hip arthroplasty. Clin Orthop Relat Res. 1981;159: 183-5.
8. . Late fracture of the acetabulum after total hip replacement. J Bone Joint Surg Br. 1972;54: 600-6.
9. , Lewallen DG, Hanssen AD, Cabanela ME. Pelvic discontinuity in revision total hip arthroplasty. J Bone Joint Surg Am. 1999;81: 1692-702.
10. , Salvati EA, Robinson HJ. Mechanical failures in total hip replacement requiring reoperation. J Bone Joint Surg Am. 1979;61: 28-36.
11. , Saluja R, Diener T. Reconstruction of major column defects and pelvic discontinuity in revision total hip arthroplasty. J Arthroplasty. 2000;15: 849-57.
12. , Garvin KL. Periprosthetic acetabular stress fracture causing pelvic discontinuity. Orthopedics. 2002;25: 83-5.
Copyright 2005 by The Journal of Bone and Joint Surgery, Incorporated