Periprosthetic osteolysis is a common cause for revision of THA. Revision is indicated when bone loss compromises implant stability or when patients have symptomatic polyethylene wear.2,4,12,13 Treatment entails exchange of the bearing surface to decrease debris production and, when possible, débridement of the lesion cavity with grafting.2,4,12,13 Some authors have advocated acetabular component removal,6,14 whereas others believe a stable acetabular component should be left in place.1,11,12 When the component is left in place, access to the osteolytic lesion for débridement and grafting is more difficult. Previous studies suggest some lesions remain untreated; 67% to 74% of osteolytic lesions are treated with grafting when the acetabular component is maintained.11,16 At short-term followup this approach reportedly provides radiographic improvement or a halting of the osteolytic lesion growth.2,11,12,16
However, radiographic evaluation of osteolysis tends to underestimate the extent of bone loss.7,8,15,17 Recently, computed tomography (CT) scans have provided a more sensitive way to evaluate the location and volume of osteolysis in the pelvis.8 CT studies suggest there are often lesions undetected with radiographs. It is, therefore, conceivable that previous studies describing osteolytic lesion treatment, which were evaluated preoperatively and post-operatively with radiographs, misrepresented the quality of the surgical technique.
We used preoperative CT to determine the number, size, and location of periacetabular osteolytic lesions present in a group of patients scheduled for reoperation on progressively enlarging or symptomatic osteolytic lesions. At the time of surgery, the osteolytic lesions were débrided to the best of the surgeon's ability and injected or packed with a calcium-sulfate-based bone graft substitute. A postoperative CT scan determined how well each osteolytic lesion was filled.
In addition to describing the preoperative osteolytic lesions, we asked four questions about the surgical technique. The first question was how many of the lesions were not treated and why were they not treated. The second and most important question was how well were the osteolytic lesions filled with graft material? The final two questions addressed the role of lesion location and method filling on the quality of lesion fill.
MATERIALS AND METHODS
We retrospectively reviewed the charts of 13 consecutive patients who had a polyethylene liner exchange and periacetabular osteolytic lesion grafting using a calcium-sulfate-based bone graft substitute (MIIG® X3 HiVisc; Wright Medical Technologies, Memphis, TN). All CT scans were obtained using a multidetector machine. Helical scans were taken at 140 kV and axial slices were reconstructed. DICOM-formatted images were saved to a compact disc and analyzed on a personal computer (Dell, Round Rock, TX) using a previously validated postprocessing software program (Muscular-Skeleton Analysis Software, Version 2.1; VirtualScopics, Rochester, NY).3,10 A single reviewer (HE) analyzed the CT data and identified the number, location, and volume of periacetabular osteolytic lesions. Each patient had a preoperative CT scan before the procedure and a postoperative CT scan the day after surgery. The same reviewer measured the percentage of the osteolytic lesion that was filled with the material on the postoperative CT scan. This retrospective study was approved by the Institutional Review Board.
There were 10 men and three women. The mean age at the time of revision was 56 years (range, 48-78 years). The original diagnosis was osteoarthritis in seven patients, developmental dysplasia in three, avascular necrosis in two, and posttraumatic osteoarthritis in one. The index revision was the first revision on the involved hip for 12 of the 13 surgeries. One patient was having his second polyethylene exchange.
The mean time from the last hip surgery until the index revision was 131 months (range, 69-175 months). All patients had radiographs demonstrating bone ingrowth of the cementless femoral and cementless acetabular components. All components were DePuy modular allowing both head and liner exchange. Eleven of the acetabular components had a single-dome hole. The remaining two components had multiple cavitary holes in the shell. Two cups required custom replacement liners because they are no longer manufactured.
The indication for surgery for these patients was impending wear-through of the polyethylene liner, rapidly progressive osteolysis, or symptoms of polyethylene wear such as hip subluxation or groin pain. In every case, the surgical approach was the same as used for the primary procedure. Ten patients had their revision surgery performed with the posterior approach and three with a modified Harding approach. At surgery, osteolytic lesions were débrided to the best of the surgeon's ability using instruments available from a MIIG® HV Procedure Kit (Wright Medical Technologies, Memphis, TN). The instruments included a malleable bristle brush, curette, and suction device. When treated, rim lesions were curetted under direct visualization and either packed by hand or the material was injected while sealing the injection portal by hand. Dome lesions were curetted and brushed using access for the brush through holes in the acetabular component or through a bone window. Dome lesions were filled by injecting material through either a bone window or a tunnel drilled into the lesion. Fluid flow into the lesion was confirmed by injecting saline into the hole or tunnel and observing runoff through the dome hole. Alternatively, dome lesions were filled by injecting through a hole in the cup without a secondary bone window or tunnel. Every attempt was made to maximize the amount of lesion filled. In every case, the material injected or packed was either calcium sulfate (MIIG® X3 HiVisc) or a combination of calcium sulfate and calcium phosphate (ProDense; Wright Medical Technologies). These materials came as a powder and a liquid, which were mixed at the time of surgery to yield either 15 or 25 mL of viscous liquid. Initially, the combination can be injected as a high-viscosity liquid. Over several minutes, the material hardens to a puttylike consistency that can be packed; eventually, it solidifies with a compressive strength comparable to bone cement. The material was chosen because of its ease of use, the availability of sufficient volume to fill large defects, its radiographic visibility, and its ability to flow during filling with subsequent hardening that afforded the potential to seal the communication pathways between the osteolytic lesion and the joint space while ensuring the graft material remained fixed until it resorbed. In four rim lesions, the material was allowed to set up until it was in a doughy phase and the lesion was hand packed.
Differences in the percentage of the lesion fill based on the presence of a cortical window, the type of lesion, and the fill technique were examined with a Mann-Whitney U test. Analysis was performed using SPSS (Statistical Package for the Social Sciences, Chicago, IL). p values of 0.05 and less were considered significant.
In the 13 patients, 22 periacetabular osteolytic lesions were identified on the preoperative CT scan (Table 1). The mean lesion volume was 20.3 mL (range, 2.4-68.1 mL). Five patients had a single osteolytic lesion, seven patients had two lesions, and one patient had three lesions. Eleven of the lesions originated at the rim of the acetabular component and 11 originated through the holes in the shell of the acetabular component.
At the time of surgery, four lesions in four patients were not treated. The mean size of these lesions was 12.2 mL (range, 2.9-26.5 mL). Three of the untreated lesions were anterior rim lesions and one was located at a posterior screw hole. In each case, the patient had more than one periacetabular lesion. Two of the rim lesions were intentionally neglected because they were less than 10 mL in volume and considered clinically unimportant. The remaining lesions were 26.5 mL and 12.9 mL. In each case, the surgeon thought a single larger lesion with interconnected lobes existed. However, at surgery, the lesion behaved as two discrete lesions either because the connection did not exist or was not fully débrided.
Eighteen of the 22 lesions were treated with curettage and placement of the bone graft substitute. The mean size of these lesions was 22.1 mL (range, 2.4-68.1 mL). Eight of the lesions originated at the rim, nine originated from the single apical dome hole, and one originated from the screw hole of a multihole cup.
On the postoperative CT, the graft material was clearly visible and the percentage of volumetric lesion fill was calculated (Fig 1). For all 18 treated lesions, the mean percentage of lesion filling was 49% (range, 0-83%). In nine lesions, the graft material filled more than 50% of the lesion. For the nine lesions in which the material filled less than 50% of the lesion, the primary explanation for poor filling was leakage or extravasation of the material through a cortical window, a screw hole, or another bone defect such as a medial wall osteolytic defect. The one lesion that had 0% filling had a 5.9-mL lesion that we attempted to fill with a tunnel into the lesion. The tunnel was approximately 2 cm long. The postoperative CT scan showed filling of the tunnel but no filling of the lesion.
Rim and dome defects demonstrated similar fill with a mean 47% and 50% filling of the defect, respectively. In five cases, the lesion was accessed through a combination of a cortical window in the ilium and either a shell hole or the acetabular rim. Despite the improved access created with the cortical window, the mean percent filling was 46% (range, 0-76%). We observed no difference between these five lesions and the 13 that were filled without a cortical window in which the mean lesion fill was 50% of the defect (range, 17-83%).
We observed no difference in filling by hand packing and injection. In the four rim lesions in which the material was allowed to set until it was in a doughy phase and the lesion hand packed, 60% of the lesion was filled on average (range, 29-80%). In the 14 lesions treated by injecting the material when it was in a liquid phase the mean filling was 46% of the lesion volume (range, 0-83%).
A debate continues on whether to remove a stable cementless acetabular component when treating pelvic osteolysis. 1,9,12 Proponents of cup removal emphasize the ease of curettage and grafting of the osteolytic lesion.5,6,14 In these cases, all granuloma is removed and the defect can be completely filled with graft. However, the procedure takes longer and there is more blood loss than when a stable acetabular component is retained.16 There is also a concern the new acetabular component will not be stable, especially if bone loss is extensive.13 Proponents of cup retention emphasize a shorter procedure and good short-term results whether or not the osteolytic defect is grafted.1,11,16 However, access to the osteolytic lesion for curettage and grafting is more difficult when the acetabular component is retained.
Ours is a retrospective analysis of 13 cases. Because the acetabular cup was retained, it was not possible to visualize the osteolytic lesion while débriding it. Consequently, we could not quantify the lesion débridement. If granuloma tissue remained or residual fluid became entrapped during grafting, it would have been impossible to obtain complete defect filling. Additionally, the complex, multilobed geometry of the large defects we attempted to fill likely contributed to our inconsistent results.
Another concern is the clinical importance of knowing the percentage of osteolytic lesion fill. The literature clearly indicates acetabular components retained in the face of osteolysis remain stable and osteolytic lesions do not progress.1,11,12,16 However, the length of followup in these reports is generally less than 10 years. It is conceivable osteolysis will not return until 5 or 10 years after revision. It is also possible the return of osteolysis may be related to the presence of graft material and the amount of graft material placed. Furthermore, the introduction and evaluation of new materials used to treat osteolytic lesions will likely be dependent on both the properties of the material and the quality of the surgical technique. This study provides initial data on the quality of the surgical technique using one material.
We favor retention of stable acetabular components and curettage and grafting when possible. We acknowledge grafting of osteolytic lesions, while maintaining a stable cementless component, is difficult. Even with preoperative CT reconstructions and surgical planning, four of 22 lesions were neglected at the time of surgery. In the 18 lesions that were treated, we were able to fill an average of 49% of the lesion volume.
Analysis of the neglected lesions reveals three of the four lesions were anterior rim lesions that are more difficult to access than dome lesions or posterior rim lesions. Two of the neglected lesions were smaller than 10 mL and were judged of minimal clinical importance. The existence of neglected or untreated lesions is not new. Schmalzried et al16 reported on 15 hips in which pelvic osteolysis was treated with acetabular component retention. In that series, 12 of the 18 lesions were grafted. Likewise, Maloney et al11 reported 74% (34 of 46) osteolytic lesions and 35 patients were grafted. The treatment of 18 of 22 lesions (82%) in our study appears better than the treatment rates in these studies. However, clearly there is room for improvement in the surgical technique.
Although on average we only filled 49% of the osteolytic lesion volume, we believe the data are important. This is the first report that has attempted to quantify the ability to fill osteolytic lesions. More importantly, it emphasizes the need to further refine the surgical technique. There were no identifiable trends related to lesion type (dome or rim), access technique (window or no window), or fill technique (injection or packing); however, the authors can speculate about ways to improve the technique. Visibility of the osteolytic lesion is difficult when the cup is retained; therefore, development of small flexible scopes offers potential to improve the technique. We also had difficulty physically débriding the lesions. Although customized tools are needed, another idea would be to use an injectable material that was capable of chemical or enzymatic débridement of the lesions. The optimum graft material will combine good biologic and physical characteristics at an affordable cost. Although an osteoinductive and osteoconductive material would be biologically ideal, further investigation is needed to determine the proper dose and carrier for delivery into large osteolytic lesions. The physical characteristics are easier to define and might differ depending on the lesion. For instance, a liquid can be injected into contained lobular lesions, whereas putty can be hand packed into uncontained peripheral lesions. Like in this study, the material should be visible with current imaging techniques so the quality of the surgery can be graded. Although refinements to the technique and material need further evaluation, longer followup with repeat CT scans or other imaging techniques will ultimately determine how the percentage of lesion fill influences clinical outcome.
We thank Serena B. Leung who assisted with compiling the data for this study.
1. Beaulé PE, LeDuff MJ, Dorey FJ, Amstutz HC. Fate of cementlessacetabular components retained during revision total hip arthroplasty. J Bone Joint Surg Am
2. Chiang PP, Burke DW, Freiberg AA, Rubash HE. Osteolysis of thepelvis: evaluation and treatment. Clin Orthop Relat Res
3. Claus AM, Totterman SM, Sychterz CJ, Tamez-Pena JG, LooneyRJ, Engh CA Sr. Computed tomography to assess pelvis osteolysisafter total hip replacement. Clin Orthop Relat Res
4. Claus AM, Walde TA, Leung SB, Wolf RL, Engh CA Sr. Management of patients with acetabular socket wear and pelvic osteolysis. J Arthroplasty
. 2003;18(suppl 1):112-117.
5. Etienne G, Bezwada HP, Hungerford DS, Mont MA. The incorporation of morsellized bone grafts in cementless acetabular revisions. Clin Orthop Relat Res
6. Etienne G, Ragland PS, Mont MA. Use of cancellous bone chips and demineralized bone matrix in the treatment of acetabular osteolysis: preliminary 2-year follow-up. Orthopedics
7. Kitamura N, Leung SB, Engh CA Sr. Characteristics of pelvic osteolysis on computed tomography after total hip arthroplasty. ClinOrthop Relat Res
8. Kitamura N, Pappedemos PC, Duffy PR, Stepniewski AS, HopperRH, Engh CA Jr, Engh CA Sr. The value of anteroposterior pelvicradiographs for evaluating pelvis osteolysis. Clin Orthop Relat Res
9. Leopold SS, Jacobs JJ, Rosenberg AG. Cancellous allograft in revision total hip arthroplasty: a clinical review. Clin Orthop RelatRes
10. Leung S, Naudie D, Kitamura N, Walde T, Engh CA. Computedtomography in the assessment of periacetabular osteolysis. J BoneJoint Surg Am
11. Maloney WJ, Herzwurm P, Paprosky W, Rubash HE, Engh CA. Treatment of pelvic osteolysis with a stable acetabular componentinserted without cement as part of a total hip replacement. J BoneJoint Surg Am
12. Maloney WJ, Paprosky W, Engh CA, Rubash H. Surgical treatmentof pelvic osteolysis. Clin Orthop Relat Res
13. Naudie DD, Engh CA Sr. Surgical management of polyethylenewear and pelvic osteolysis with modular uncemented acetabularcomponents. J Arthroplasty
. 2004;19(suppl 1):124-129.
14. Peters CL, Erickson JA, Dunn HK. Revision of well-fixed cementless acetabular components for polyethylene failure. Clin OrthopRelat Res
15. Puri L, Wixson RL, Stern SH, Kohli J, Hendrix RW, Stulberg SD. Use of helical computed tomography for the assessment of acetabular osteolysis after total hip arthroplasty. J Bone Joint Surg Am
16. Schmalzried TP, Fowble VA, Amstutz HC. The fate of pelvic osteolysis after reoperations: no recurrence with lesional treatment. Clin Orthop Relat Res
17. Stulberg SD, Wixson RL, Adams AD, Hendrix RW, Bernfield JB. Monitoring pelvis osteolysis following total hip replacement surgery: an algorithm for surveillance. J Bone Joint Surg Am