Clinical and Radiological Results of Extended Trochanteric Osteotomies versus Intraoperative Femoral Fracturesin Revision Hip Arthroplasty

 Matthias Lerch*,Henning Windhagen*,Gabriela von Lewinski*,Fritz Thorey*

Abstract:

Removal of the femoral stem can be challenging for the orthopaedic surgeon. The decision between performing an extended trochanteric osteotomy (ETO) or trying to remove the femoral stem without an osteotomy and taking the risk of an intraoperative fracture is often hard to make.

Purpose of the study was thus to describe our experiences, comparing intraoperative femoral fractures during stem removal with ETOs in femoral revision arthroplasties.

45 intraoperative fractures during revision hip arthroplasty were compared to a collective of 28 femoral revision arthroplasties. Preoperatively and after 32 months (range, 21.6 - 76 months) the patients were examined clinically and radiographically. The SF-36 health score, Harris hip score, a motion- and pain score and the occurrence of postoperative complications were collected.

Only in ETO patients the Harris hip score increased significantly (p > 0.01). Results in pain- and motion scores were better in the osteotomy group. Outcomes in the SF-36 health score in 3 dimensions were significantly better in patients with ETOs. No osteosynthesis related complications occurred in the ETO group, but in 6 (13.3 %) patients in the fracture group (p > 0.01). Re-revision rate was lower in the osteotomy group. Radiographs showed less bone resorption, decreased stem subsidence and fewer nonunion in patients with ETOs.

Well conducted ETOs are preferable to unplanned femoral fractures during stem removal.

J.Orthopaedics 2007;4(4)e6
 index.htm
Introduction:

The number of femoral revision arthroplasties continues to rise in an ageing population [9]. Removal of the femoral stem can be a challenge for an orthopaedic surgeon and the decision between performing a femoral osteotomy or taking the risk of an intraoperative fracture, being aware that the occurrence of this complication may lead to disastrous results, is often hard to make. High fracture rates reported, point out that stem removal is associated with substantial surgical complications [8, 19]. An intraoperative fracture rate of 7.8 % was reported in revision Total Hip Arthroplasties (THA) in an review of the Mayo Clinic Joint Registry [3]. In another study, intraoperative femoral fractures only occurred in revision hip arthroplasties at the author’s centre [10]. Christensen et al. reported of 10 fractures which occurred exclusively during 159 revision surgeries (6.3 %) that showed fracture union, but only 6 of 10 had satisfactory postoperative function [6].
Preserving as much bone stock as possible and avoiding any femoral fracture or crack is crucial in femoral revision procedure [8]. Extended trochanteric osteotomy (ETO) can contribute to this, by facilitating cement and stem removal or implantation of revision components [4, 8, 14, 25, 26]. Additionally, this technique may minimize the risk of intraoperative fractures and preserves the remaining femoral bone stock [8, 14, 25, 26]. However, it has also been associated with new hazards like increased incidence of nonunion, fracture of the osteotomy fragment and subsidence of the stem [4, 19, 26].
To our knowledge it is not yet clear if a permissive indication for ETO may lead to better postoperative results than removing stem and cement from the top of the femur eventually causing fractures.
Purpose of the present study was thus to describe our experiences, comparing intraoperative femoral fractures during stem removal with extended trochanteric osteotomies in femoral revision arthroplasties in a retroperspective clinical and radiological study.

Material and Methods :

Between January 1992 and February 2004 593 femoral revision arthroplasties were performed at our orthopaedic department. All surgical procedures were carried out by a group of four experienced surgeons. Out of this number of revision arthroplasties two groups were collected for the present study. 45 intraoperative fractures were detected (group I) and compared to a collective of 28 patients which underwent an ETO during femoral revision arthroplasty (group II). Preoperatively and after a follow-up period of 32 months (range, 21.6 months to 76 months) patients were examined clinically and radiographically.
There were 32 (71.1 %) female and 13 (28.9 %) male patients with an age range of 38.5 years – 88.2 years (mean 67.5 years) in the fracture group and 10 (35.7%) female and 18 (64.3%) male patients with an age range of 36.7 years – 83.8 years (mean 65.6 years) in the osteotomy group. No statistically significant differences regarding age at operation were apparent between the groups. The indication for revision surgery was aseptic loosening for 37 hips (82.2 %) in the fracture group and for 21 hips (75.0 %) in the osteotomy group, septic loosening for 5 hips (11.1 %) in the fracture group and for 6 hips (21.4 %) in the osteotomy group and recurrent luxations for 3 hips (6.7%) in the fracture group and for 1 hip (3.6%) in the osteotomy group. Patients with ETOs had 1.7 (range, 1 to 4) and in the fracture group 1.6 (range, 1 to 5) previous surgeries on the involved hip. No statistically significant differences in indication for revision surgery and number of previous surgeries were apparent between the groups.
Classification of fractures:
In group I, fractures during femoral revision arthroplasties occurred during dislocation, stem or cement removal, instrumentation or canal preparation. Fractures were graded using the Vancouver Classification adapted to the intraoperative scenario [12].
7 (15.5 %) displaced or unstable fractures of the greater trochanter (A3), 15 (33.3 %) diaphyseal cortical perforations (B1), 9 (20 %) undisplaced linear cracks (B2) and 7 (15.5 %) displaced fractures of the midfemur (B3) were monitored. Furthermore, 5 (11.1 %) displaced fractures of the distal femur (C3), 1 (2.2 %) cortical perforation of the distal femur (C1) and 1 (2.2 %) undisplaced linear crack propagating to the distal femur (C2) occurred. Cerclage wire osteosynthesis was performed in 34 (75.5 %) hips. For 8 (17.8 %) femoral fractures no osteosynthesis was needed. In 3 hips (6.7 %) a combination of plates and cables was brought in. In the osteotomy group, fragments were reattached with at least 2 cables.
Surgical technique:
We used a standard lateral approach for exposure of the femur and acetabulum that was extended distally as far as necessary for the ETO and in case of intraoperative fractures. The ETO first described by Wagner [22] and then modified by Younger et al. [25] was performed in all cases in the osteotomy group with the femoral component still in place. The decision making process for the ETO involved patient age and condition, implant design, the use of cement and preoperative radiographs. However, the final decision was made intraoperatively after the surgeon assessed bone stock, soft tissue and implant stability.
With an oscillating saw the posterolateral third of the femur including the complete greater trochanter was cut, beginning with the anterior lateral osteotomy. Next, the distal end of the osteotomy was performed with the use of a pencil-tip burr to minimise stress risers and the risk of fractures. After the posterior cut, the fragment was retracted anteriorly (Fig. 1).
Only for removal of cemented stems, the distal osteotomy site was positioned beyond the tip of the stem, in order to facilitate cement removal. The osteotomy fragment was retracted always with the attached vastus lateralis and abductors to retain blood supply and innervation. Bone grafts were used for osteotomy reinforcement in 7 (25.0 %) cases and for 10 (22.2 %) fractured femurs. All patients received a dull corundum coated uncemented titanium-aluminium-niobium alloy long-stem femoral component (Wagner SL revision stem; Protek AG, Bern, Switzerland) except for 5 (17.9%) patients in the osteotomy group and 10 (22.2%) patients in the fracture group who received a BiCONTACT® Stem (Aesculap, Tuttlingen, Germany) originally designed for primary THA. In every instance the distal end of the osteotomy or fracture was bypassed by at least two femoral diameters [12, 22].

Figure 1 – Schematic view on the prosthesis after extended trochanteric osteotomy. Proximal abductor musculature and vastus lateralis (not pictured) remain attached to the fragment.

Figure 2 – Improvements of the SF-36 health score in patients with ETOs and in patients with fractures during femoral revision arthroplasties.

Figure 3a – Three-week postrevision radiograph of a 73-year-old woman shows distal end of the ETO, a well adapted osteotomy fragment and good alignment. Figure 3b – Follow-up radiograph 6 years postoperatively shows healed osteotomy without migration and the ingrown femoral stem in good position

Postoperative treatment:
Depending on the femoral component, acetabular reconstruction, stability of the hip, and the patients’ bonestock quality, postoperative treatment varied. After toe-touch weightbearing was maintained for 6 weeks, weightbearing was increased to full bodyweight by 12 weeks. For 6 to 12 weeks, flexion of the hip was limited to 90°, and active abduction and external rotation were avoided. Strengthening exercises and passive range of motion were begun on postoperative day 2.
Preoperatively and after a follow-up period of 32 months (range, 21.6 months to 76 months) patients were examined clinically and radiographically.
Clinical evaluation:
The patients were routinely followed-up at 3 months, 12 months, 36 months and in intervals of 3 years in our outpatient clinic. Furthermore patient records were included. For the clinical follow-up examination the Harris hip score, motion score (visual analogue scale for flexion of the hip 0 – 6 points) and pain score (visual analogue scale 1 – 10 points) were used. The SF-36 health score (QualityMetric, Incorporated, Lincoln, RI.) was employed to assess subjective health improvement including 8 multi-item scales: physical functioning, role physical, bodily pain, general health, vitality, social functioning, role emotional and mental health, ranging from 0 (maximal symptoms and poor health) to 100 (no symptoms and excellent health) [23].
In addition, the occurrences of limping and postoperative complications were assessed.
Radiographic evaluation:
Preoperatively and at follow-up ante-posterior (ap) and axial radiographs were taken. In 6 patients 3 year follow-up radiographs were not available. Radiographic evaluation included the occurrence of fractures or nonunions. Stem migration was evaluated by measuring the vertical distance between the shoulder of the prosthesis and the tip of the greater trochanter on the immediate postoperative ap radiographs and at final follow-up. Migration of the osteotomy fragment was measured as the distance between the trochanter fragment and the host bone on ap radiographs [5]. Since trochanteric fragments may show migration to craniomedial, axial radiographs were assessed to quantify the correct distance whenever possible. Implant alignment was measured by a stem deviation of > 3° from the femoral longitudinal axis. Osteosynthesis failure, bone resorption according to Gruen et al. [7] and bone quality (osteoporosis) were assessed on all radiographs. Bone resorption affecting less than 50 % of 1 Gruen Zone was rated as not relevant. The criterion for osteoporotic conditions was, when the total cortical thickness was less than 25 % of the total femoral calibre at the midpoint of the shaft [24].
Statistical analysis:
Pooled data of both groups were analysed using 2-sided Student’s t-test, paired t-test and Pearsons Chi-Square-Test using SPSS (version 11.5; Chicago, Illinois). A p value < 0.05 was considered significant.

Results :

6 patients in the fracture group died 2.6 years after the operation. No patient in the ETO group was lost to follow-up.
Clinical results:
Results showed an increase in Harris hip scores, pain and motion scores for all 73 femoral revision arthroplasties. Only patients with ETOs showed a significant increase in Harris hip scores (Tab. 1).

Table 1. Clinical scores: Extended trochanteric osteotomies and femoral fractures in femoral revision arthroplasty

In all dimensions of the SF-36 health score, patients with ETOs had better improvements compared to patients with intraoperative femoral fractures. Improvements in dimensions “Physical Functioning”, “Bodily Pain” and “Vitality” were significantly higher in the osteotomy group (Fig. 2).
Joint luxation occurred in 3 (6.7 %) patients with intraoperative fractures and once (3.6 %) in the osteotomy group (not significantly different by Chi-Square-Test). After closed reduction and change of the femoral head postoperatively, no more luxations occurred in these patients. No osteosynthesis related complication was seen in the osteotomy group, but in 6 (13.3 %) patients with intraoperative femoral fractures (p < 0.01): 1 (2.2 %) patient complained about prolonged postoperative pain over the implanted hardware, in 2 (4.4 %) cases cable failure occurred during mobilisation, 1 patient needed refixation of the greater trochanter because of craniomedial migration and in 2 cases osteosynthesis revision was needed because of painful hardware. After ETO, 1 (3.6 %) patient had a transitory lesion of the femoral nerve and 2 (7.1 %) patients had mild Trendelenburg gait, although no trochanteric nonunion in follow-up radiographs was found. 3 (8.6 %) hips were re-revised in the fracture group (1 nonunion and subsequent late periprosthetic femoral fracture, 1 aseptic and 1 septic loosening due to nonunion) and 1 (3.6 %) hip needed re-revision in the osteotomy group (septic loosening and nonunion).
Radiological results:
Radiographs showed better results in patients with ETOs (Tab. 2), though they were not significantly different by Chi-Square-Test.

Table 2. Radiographic results: Extended trochanteric osteotomies and femoral fractures in femoral revision arthroplasty

In all patients except for 1 (3.6 %) in patients with ETOs and 3 (8.6 %) in patients with intraoperative femoral fractures, osteotomies and fractures achieved union (Fig. 3).  

Every implanted stem had excellent alignment (< 3°) in follow-up radiographs. The number of patients with osteoporotic conditions was similar with 26 (58 %) patients in the fracture group and 16 (57 %) in ETO patients. In the fracture group, periprosthetic bone resorption was found in 4 cases: in 1 (2.2 %) case Gruen Zones 1, 2, 7, 8, 9, and 14 were affected, 1 femur showed bone resorption in Gruen Zone 1, 1 femur in Zone 5 and in 1 hip Gruen Zones 2 and 6 were affected. In the osteotomy group, only Gruen Zone 7 was affected in 1 (3.6 %) patient.

Discussion:

Femoral osteotomies during femoral revision arthroplasty have always been discussed controversially. Many [8, 14, 25, 26] support the use of the ETO, although some authors have concerns with this technique [18, 21] and even give advices to avoid any femoral osteotomy [15]. However, it is not yet clear if performing an ETO is preferable to removing the femoral stem without an additional osteotomy and taking the risk of an intraoperative fracture.
Purpose of the study was thus to describe our experiences, comparing intraoperative femoral fractures during stem removal with ETOs in femoral revision arthroplasties.

In this study, all 73 femoral revision arthroplasties had good clinical results in Harris hip scores pain- and motion scores, but solely in patients with femoral osteotomies the increase in Harris hip score was significant. Results in visual analogue scale for motion and pain were better in the osteotomy group. Subjective patient physical and psychological satisfaction with the operation, represented by the SF-36 health score, showed a higher increase in all dimensions in patients with ETOs. However, significantly better increases were only seen in ”Physical Functioning” and “Bodily Pain”, as physical dimensions, and “Vitality”, as a psychological dimension.
The rate of osteosynthesis related complications was significantly higher (p < 0.01) in patients with intraoperative fractures. Furthermore joint luxation was uncommon in patients with ETOs, whereas it complicated rehabilitation in 3 patients in the fracture group. In the osteotomy group 2 patients with Trendelenburg gait were found. Since we did not find a trochanteric nonunion in these patients, we concur with Nicholson et al., suggesting a neuromuscular cause [17].
Radiographs showed better results in patients with ETOs. Periprosthetic bone resorption was found in 10 Gruen Zones in the fracture group, but only 1 Zone in the ETO group was affected. Although the overall results in the ETO group were better compared to the fracture group, we found only few statistically significant differences between the groups. Regarding the high number of only minor fractures, the outcome observed in the fracture group appears to be even poorer compared to the ETO group. This may demonstrates that small cracks or perforations during femoral revision might sometimes be underestimated, especially when previous operations and poor bonestock make the patient at risk and the bone healing capacity is compromised. In our opinion undisplaced linear cracks are easily fixed intraoperatively, but seem to jeopardize prosthesis ingrowth and stability postoperatively, since these types of cracks might propagate undetected. However, further studies have to be conducted to find out why cracks in some patients do finally become stable and in some they do not.

We note some limitations to our study. Besides the limitations that come along with a retrospective work, 6 patients in the fracture group were lost for follow-up after 2.6 years.
We used the SF-36 health score in our study although it is not a common evaluation tool in orthopaedic surgery, therefore we were not able to discuss the results of the SF-36 health score with the literature. However, we believe that we found an acknowledged and reliable instrument to reflect both physical and psychological effects [1] in patients after femoral revision arthroplasties.
Though the two groups are statistically similar, they may have obtained different treatments intraoperatively. The groups are possibly different in bonestock damage or in other issues that led the surgeon to the decision to perform the ETO or to remove the stem from the top of the femur. Furthermore, it is unknown if all femurs in the ETO group would have fractured spontaneously or not. Nevertheless, we believe that the collectives presented here are well comparable, since the number of patients with osteoporotic conditions was similar and indications for revision surgery and age at revision surgery showed no statistical differences. Even the number of previous operations, which may be considered as one of the most determined factors for success in revision surgery, was similar.

This study reflects the problem of intraoperative decision making in removal of well fixed stems. On the one hand, removing stem and cement from the top of the femur would be eligible, but the surgeon has to take the risk of an incalculable fracture. On the other hand, performing an ETO facilitates stem removal and may minimize the risk of intraoperative fractures, though it is an iatrogenic but controllable bone trauma. Addressing this controversial issue, our results suggest that a more permissive indication for femoral osteotomy may lead to better postoperative results, especially in patients with poor bone stock. Other authors, however, have concerns with the ETO. They believe that in femoral revision surgery any weakening of the femur, both intentionally or by accident, compromises femoral bonestock and may debase postoperative results [21]. In a recently conducted in vitro cadaver study, “the ETO resulted in a significant reduction of torsional strength and energy required for fracture” [18]. The authors suggest that rehabilitation should be even more restrictive after revision total hip arthroplasty with use of the ETO. One author observed femoral fractures during femoral revision arthroplasty with the use of the ETO in 12 % (5 of 43 hips) [8]. In a larger series, 25 (20 %) iatrogenic fractures of 122 revision surgeries were found when this technique was performed [2]. A fracture rate of 10.8 % (18 of 166 hip revisions), propagating from the distal end of the osteotomy site, were reported in a further study [14]. Nevertheless, removal of cement, broken implants or well fixed stems without gaining proper exposure may jeopardize the femurs integrity, leading to uncontrolled fractures sometimes even propagating into the supracondylar region [20]. Especially in patients with poor bone stock, osteoporotic conditions and previous operations or in septic revision surgery, compromising the femoral cortex may likely happen. In our clinical experience, osteosynthesis of an uncontrolled fracture in these patients poses a great challenge to the surgeon.
The results found in patients with ETOs in our series concur well with the findings of other authors [2, 8, 14, 25, 26]. Several authors [13, 16] have published techniques to address the issues of safe cement and stem removal. Wagner [22] described an osteotomy technique were the greater trochanter and half of the circumference of the femoral cortex is included. Popularized by Younger et al. [26], this technique provides excellent implant, fragment and cement exposure, correction of proximal femoral deformity, neutral reaming and excellent acetabular exposure [8, 14, 25, 26]. Besides the excellent results reported by Paprosky et al. [14, 25, 26] and recently by Mardones et al. [11] including low complication rates in fractures, fragment migration, infection, stem subsidence, instability of the stem and nonunion, we found comparably low complication rates for stem subsidence and luxation, no osteosynthesis failure, an acceptable union rate, decreased postoperative pain and increased range of motion.

Conclusion:

Despite the critical reports found in literature, this study demonstrates that well conducted extended femoral osteotomies are preferable to unplanned and uncontrollable femoral fractures during stem removal.

References : 

1.  Adachi JD, Loannidis G, Berger C, Joseph L, Papaioannou A, Pickard L, Papadimitropoulos EA, Hopman W, Poliquin S, Prior JC, Hanley DA, Olszynski WP, Anastassiades T, Brown JP, Murray T, Jackson SA, Tenenhouse A (2001) The influence of osteoporotic fractures on health-related quality of life in community-dwelling men and women across Canada. Osteoporos Int 12: 903-908

2. Aribindi R, Paprosky W, Nourbash P, Kronick J, Barba M (1999) Extended proximal femoral osteotomy. Instr Course Lect 48: 19-26

3.  Berry DJ (1999) Epidemiology: hip and knee. Orthop Clin North Am 30: 183-190

4. Chen WM, McAuley JP, Engh CA, Jr., Hopper RH, Jr., Engh CA (2000) Extended slide trochanteric osteotomy for revision total hip arthroplasty. J Bone Joint Surg Am 82: 1215-1219

5.  Chin KR, Brick GW (2000) Reattachment of the migrated ununited greater trochanter after revision hip arthroplasty: the abductor slide technique. A review of four cases. J Bone Joint Surg Am 82: 401-408

6. Christensen CM, Seger BM, Schultz RB (1989) Management of intraoperative femur fractures associated with revision hip arthroplasty. Clin Orthop Relat Res: 177-180

7. Gruen TA, McNeice GM, Amstutz HC (1979) "Modes of failure" of cemented stem-type femoral components: a radiographic analysis of loosening. Clin Orthop Relat Res: 17-27

8. Huffman GR, Ries MD (2003) Combined vertical and horizontal cable fixation of an extended trochanteric osteotomy site. J Bone Joint Surg Am 85-A: 273-277

9.  Huo MH, Cook SM (2001) What's new in hip arthroplasty. J Bone Joint Surg Am 83-A: 1598-1610

10. Johansson JE, McBroom R, Barrington TW, Hunter GA (1981) Fracture of the ipsilateral femur in patients wih total hip replacement. J Bone Joint Surg Am 63: 1435-1442

11. Mardones R, Gonzalez C, Cabanela ME, Trousdale RT, Berry DJ (2005) Extended femoral osteotomy for revision of hip arthroplasty: results and complications. J Arthroplasty 20: 79-83

12. Masri BA, Meek RM, Duncan CP (2004) Periprosthetic fractures evaluation and treatment. Clin Orthop Relat Res: 80-95

13.  Masterson EL, Masri BA, Duncan CP (1998) Surgical approaches in revision hip replacement. J Am Acad Orthop Surg 6: 84-92

14.  Miner TM, Momberger NG, Chong D, Paprosky WL (2001) The extended trochanteric osteotomy in revision hip arthroplasty: a critical review of 166 cases at mean 3-year, 9-month follow-up. J Arthroplasty 16: 188-194

15. Morrey BF, Kavanagh BF (1992) Complications with revision of the femoral component of total hip arthroplasty. Comparison between cemented and uncemented techniques. J Arthroplasty 7: 71-79

16.  Nercessian OA, Newton PM, Joshi RP, Sheikh B, Eftekhar NS (1996) Trochanteric osteotomy and wire fixation: a comparison of 2 techniques. Clin Orthop: 208-216

17.  Nicholson P, Mulcahy D, Fenelon G (2001) Trochanteric union in revision hip arthroplasty. J Arthroplasty 16: 65-69

18.  Noble AR, Branham DB, Willis MC, Owen JR, Cramer BW, Wayne JS, Jiranek WA (2005) Mechanical effects of the extended trochanteric osteotomy. J Bone Joint Surg Am 87: 521-529

19.  Paprosky WG, Martin EL (2002) Removal of well-fixed femoral and acetabular components. Am J Orthop 31: 476-478

20.  Schurman DJ, Maloney WJ (1992) Segmental cement extraction at revision total hip arthroplasty. Clin Orthop Relat Res: 158-163

21.  Turner RH, Emerson RH, Jr. (1982) Femoral Revision Total Hip Arthroplasty. In: Turner RH, Scheller AD Jr, ed. Revision Total Hip Arthroplasty. New York, Grune and Stratton: 75-104

22. Wagner H (1989) [A revision prosthesis for the hip joint]. Orthopade 18: 438-453

23.  Ware JE, Jr., Gandek B (1998) Overview of the SF-36 Health Survey and the International Quality of Life Assessment (IQOLA) Project. J Clin Epidemiol 51: 903-912

24.  Watts NB, Harris ST, Genant HK, Wasnich RD, Miller PD, Jackson RD, Licata AA, Ross P, Woodson GC, III, Yanover MJ, . (1990) Intermittent cyclical etidronate treatment of postmenopausal osteoporosis. N Engl J Med 323: 73-79

25.  Younger TI, Bradford MS, Magnus RE, Paprosky WG (1995) Extended proximal femoral osteotomy. A new technique for femoral revision arthroplasty. J Arthroplasty 10: 329-338

26.  Younger TI, Bradford MS, Paprosky WG (1995) Removal of a well-fixed cementless femoral component with an extended proximal femoral osteotomy. Contemp Orthop 30: 375-380

This is a peer reviewed paper 

Please cite as : Matthias Lerch: Clinical and Radiological Results of Extended Trochanteric Osteotomies versus Intraoperative Femoral Fracturesin Revision Hip Arthroplasty

J.Orthopaedics 2007;4(4)e6

URL: http://www.jortho.org/2007/4/4/e6

ANNOUNCEMENTS

Arthrocon 2011


Refresher Course in Hip Arthroplasty

13th March,  2011

At Malabar Palace,
Calicut, Kerala, India

Download Registration Form

For Details
Dr Anwar Marthya,
Ph:+91 9961303044

E-Mail:
anwarmh@gmail.com

Powered by
VirtualMedOnline