Sciatic Nerve Injury in Total Hip Resurfacing

Transcription

1 The Journal of Arthroplasty Vol. 25 No Sciatic Nerve Injury in Total Hip Resurfacing A Biomechanical Analysis Dustin P. Gay, MD, Dana R. Desser, MD, Brent G. Parks, MSc, and Henry R. Boucher, MD Abstract: The condition of the gluteal sling was a significant factor in determining the pressure experienced by the sciatic nerve during acetabular exposure in total hip resurfacing via a posterior approach. The position of the knee did not play a significant role at this stage of the procedure. Average pressures were not elevated above a predefined injury level during positioning for femoral preparation. During hip reduction, knee positioning seemed to play a significant role in pressures placed on the sciatic nerve. These findings suggest that releasing the gluteal sling during a posterior approach for total hip resurfacing may help to prevent postoperative sciatic nerve palsies. Consideration should also be given to at least partially flexing the knee during hip reduction in this procedure. Keywords: biomechanical, hip, total hip resurfacing, sciatic nerve Elsevier Inc. All rights reserved. Nerve palsies are a relatively uncommon but catastrophic complication of total hip arthroplasty [1-3]. Only approximately 40% [2] of patients have complete recovery of nerve palsies that result from total hip arthroplasty. Despite research and intraoperative neuromonitoring, approximately 50% of these nerve palsies are due to unknown causes [1,3]. Therefore, any intraoperative factors felt to contribute to potential nerve palsies remain important. The first new-generation metal-on-metal total hip resurfacing was recently approved for use in the United States after use outside the United States for more than 10 years [4,5]. Early US results suggest concern with complications, including nerve palsy, as compared with traditional hip arthroplasty. One US study of total hip resurfacing found a 1.7% incidence of postoperative nerve palsies at 1-year follow-up [6] as compared with an incidence of 1% in traditional total hip arthroplasty [2]. An Australian study found a 2.1% incidence of postoperative nerve palsies with modern total hip resurfacing [7]. This higher incidence of nerve palsies From the Department of Orthopaedic Surgery, The Union Memorial Hospital, Baltimore, Maryland. Submitted April 16, 2009; accepted August 26, Departmental research funds were received in support of this study. Financial disclosure: The authors state no potential conflict of interest with regard to the subject of this study. Reprint requests: Henry R. Boucher, MD, c/o Lyn Camire, Editor, Union Memorial Orthopaedics, The Johnston Professional Building, #400, 3333 North Calvert Street, Baltimore, MD Elsevier Inc. All rights reserved / $36.00/0 doi: /j.arth may become increasingly important as total hip resurfacing gains in popularity. The problem may be further exacerbated by use of this procedure in a younger, more active population in whom nerve palsies will likely result in even greater loss of function. Younger patients' heightened sense of dissatisfaction will result from their desired and expected increased activity level when compared with older patients. It is important to discover what aspects of the hip resurfacing technique might be associated with risk of nerve injury. Some investigators have suggested that the gluteal sling plays a role in sciatic nerve palsies in total hip arthroplasty. The gluteus maximus has a broad insertion. The upper fibers insert into and blend with the iliotibial band. The lower fibers, referred to as the gluteal sling, insert into the gluteal tuberosity of the femur and the lateral intermuscular septum. This occurs just distal the insertion of the quadratus femoris. At this level, the sciatic nerve lies medial and deep to the gluteal sling and lateral to the ischial tuberosity. Hip flexion and internal rotation, which occur during hip arthroplasty procedures, result in tightening of the gluteal sling and potential entrapment of the nerve against the ischial tuberosity. Hurd et al [8] have stated that an intact gluteal sling can lead to sciatic nerve compression during positioning for femoral preparation during hip arthroplasty. The position of the knee may also play a role as a result of its effects on tensioning of the sciatic nerve. Although the exact pressure required to injure a peripheral nerve in humans is not known, investigators have found endoneural edema indicating nerve injury with compressive pressures as low as 400 mm Hg in as 1295

2 1296 The Journal of Arthroplasty Vol. 25 No. 8 December 2010 little as 15 minutes using a rabbit tibial nerve model [9]. Although some portions of the total hip resurfacing procedure are transient, such as hip reduction, other portions of the procedure, such as acetabular exposure and preparation and femoral preparation, can take longer than 15 minutes. Using a pressure of 400 mm Hg or greater as an indicator of nerve injury could provide preliminary understanding of potential causes of sciatic nerve injury in humans. We hypothesized that an intact gluteal sling and an extended knee position would result in higher pressures on the sciatic nerve during acetabular exposure, positioning for femoral preparation, and hip reduction in total hip resurfacing in a cadaver model. Our purpose was to compare pressure in the sciatic nerve at these technique stages with the gluteal sling in intact and released condition and with the knee in both full extension and 90 degrees of flexion, with the posterior approach used in total hip resurfacing. Materials and Methods Ten hips from 5 fresh cadaveric lower torso specimens harvested through the thoracolumbar spine were used. There were 4 male and 1 female specimens aged 71.6 years (range, years). Each specimen was checked for previous surgical scars around the hip to ensure that no prior hip procedures had been performed. Each specimen was positioned on a peg board in a standard lateral position, and stability of the setup was confirmed by placing the operative leg through a range of motion. A standard posterior hip incision was made in a curvilinear fashion over the posterior tip of the greater trochanter. Dissection was carried down through the tensor fascia lata and gluteus maximus in a standard fashion. The greater trochanteric bursa was then resected, and the sciatic nerve was identified. The sciatic nerve was then carefully exposed both proximally and distally (Fig. 1) while leaving the short external rotators intact. The dissection was carried distally to just below the level of the gluteal sling insertion. Next, a 3- to 4-cm incision was made proximal and perpendicular to the previous incision. This incision was used to pass a calibrated 6-panel 6900 I-Scan sensor (Tekscan, Inc, South Boston, Mass). The final sensor was constructed by using 6 sensor panels, each measuring 1 1 cm, from two 4-panel 6900 I-Scan sensors. Each 6900 I-Scan sensor has 4 fingers with a 1 1-cm sensor panel on the end. These 4-sensor panels and 2-sensor panels from a second 6900 I-Scan Sensor were glued together to construct a 6-panel sensor. The 6-panel sensor covered the nerve from the ischial tuberosity to below the gluteal sling insertion. This includes the area of sciatic nerve compression injury described by Hurd et al [8] from magnetic resonance imaging evaluations. The sciatic nerve was instrumented (Fig. 2) with the I- Scan sensor. The sensor was attached to the sciatic nerve with a thin layer of glue, and secure fixation was confirmed. The interval between the abductors and external rotators was identified. The external rotators were released, and a capsulotomy was performed while Fig. 1. Illustration shows anatomic detail including the gluteal sling and ischial tuberosity.

3 Total Hip Resurfacing Sciatic Nerve Injury Gay et al 1297 Fig. 2. Photograph shows placement of sensor on sciatic nerve. The ischial tuberosity is located beneath the sciatic nerve and is covered by soft tissue in the cephalad portion of the wound. leaving the gluteal sling insertion intact. The I-Scan sensor was then attached to its base unit, which was connected to a computer with Tekscan software to collect pressure data. The hip was then dislocated posteriorly with traction and internal rotation. A pocket was created anterior and superior to the acetabulum to allow translation of the femoral head for acetabular exposure. A retractor was placed anterior to the acetabulum and used to translate the femur and femoral head anteriorly to expose the acetabulum. A second retractor was placed posteriorly. Whereas gentle retraction was performed, a small episiotomy was made inferiorly just through the capsule to allow better exposure. Completion of the acetabular exposure including labral resection was performed. All exposures were performed by the first author, who was blinded to pressure recordings throughout each procedure. Recordings were taken of the pressure experienced by the sciatic nerve with the described standard acetabular exposure necessary to prepare for acetabular reaming. Data were collected first with the knee at 90 degrees of flexion and then with the knee at 0 degree of flexion. Each position was held for approximately 5 seconds, whereas a pressure recording was taken by another investigator and the primary investigator holding the leg remained blinded to the pressure readings. The retractors were then removed. The leg was positioned for femoral preparation with hip flexion and internal rotation to center the femoral head within the wound, followed by application of axial femoral loading to deliver the femoral head out of the wound. Pressures placed on the sciatic nerve were again recorded with the knee at both 90 and 0 degrees of flexion. Pressure data were again collected by the coinvestigator, maintaining blinding of the primary investigator. The hip was then reduced with the knee in 90 degrees of flexion, followed by reduction with the knee in 0 degree of flexion using traction and external rotation. The reduction was performed in a standard fashion with gentle posterior soft-tissue retraction to prevent entrapment within the acetabulum. Again, pressure experienced by the sciatic nerve was recorded as was done previously by the coinvestigator maintaining blinding of the primary investigator. The gluteal sling insertion was then sharply released, and the hip was again posteriorly dislocated. Then the entire process of acetabular exposure, positioning for femoral preparation, and reduction was repeated as described above with consistent retractor placement. Pressure recordings were again obtained in a blinded fashion. After completion of data collection, the pressure sensor was examined to ensure that its position had remained unchanged throughout the data collection process. Statistical Analysis Power analysis of the first 4 specimens determined that 10 hips were needed to obtain a 75% ability to detect a significant pressure difference on the sciatic nerve during acetabular exposure at a significance level of.05. Two-way repeated-measures analysis of variance testing was used to compare the data. Significance was set at P.05. Results During acetabular exposure with the gluteal sling intact, the average pressure experienced by the sciatic nerve was ± mm Hg (mean ± SD) when the 2 knee position groups were combined. The average pressure with the sling intact and knee fully extended was ± mm Hg, compared with ± mm Hg with the knee in 90 degrees of flexion. This difference was not significant (P =.29). The average pressure in both groups exceeded the established critical value of 400 mm Hg. During acetabular exposure with the gluteal sling released, the average pressure exerted on the sciatic nerve, when combining the 2 knee position groups, was ± mm Hg. The average pressure with the knee fully extended was ± mm Hg, compared with ± mm Hg with the knee flexed 90 degrees (P =.44; Fig. 3). When the entire sling intact group including the knee at 0- and at 90-degree flexion was compared with the entire sling released group, the difference was significant (P =.001). When the sling intact knee at 0-degree group was compared with the sling released knee at 0-degree group, the difference was significant (P =.002). When the sling intact knee at 90-degree group was compared with the sling released knee at 90-degree group, the difference was significant (P =.003). Thus, the condition of the gluteal sling was a significant factor in determining the pressure experienced by the sciatic nerve during acetabular exposure. The position of the knee did not play a significant role.

4 1298 The Journal of Arthroplasty Vol. 25 No. 8 December 2010 Fig. 3. Graph showing comparisons during acetabular exposure. Horizontal line at 400 mm Hg shows predetermined pressure level at which nerve injury was considered possible. AE indicates acetabular exposure; SI, sling intact; SR, sling reduced; 0, knee at 0 degree of flexion; 90, knee at 90 degrees of flexion. Error bars show SD. Fig. 4. Graph showing comparisons during femoral positioning. Horizontal line at 400 mm Hg shows predetermined pressure level at which nerve injury was considered possible. FP indicates femoral positioning; SI, sling intact; SR, sling reduced; 0, knee at 0 degree of flexion; 90, knee at 90 degrees of flexion. Error bars show SD. The average pressure during positioning for femoral preparation with the gluteal sling intact, when the 0- and 90-degree flexion groups were combined was ± mm Hg, compared with 0.0 ± 0 mm Hg with the gluteal sling released (P =.09). The average pressure for the sling intact group was ± mm Hg with the knee at 0 degree of flexion, compared with ± mm Hg with the knee 90 degrees flexed (P =.07). The average pressure for the sling released group was 0.0 ± 0 mm Hg with the knee at 0 degree of flexion and 0.0 ± 0 mm Hg with the knee 90 degrees flexed (P = 1.0). There was a significant difference in pressure placed on the sciatic nerve during positioning for femoral preparation between the sling intact knee 0-degree flexed group and the sling released knee 0-degree flexed group (P =.05). There was no significant difference when the same comparison was made with the knee 90-degree flexed groups (P =.19). Therefore, during positioning for femoral preparation, only a combination of knee positioning and condition of the gluteal sling made a significant difference in the pressure experienced by the sciatic nerve. The pressures during positioning for femoral preparation did not reach the predefined critical level of 400 mm Hg (Fig. 4). Pressures experienced by the sciatic nerve during hip reduction were also recorded with the gluteal sling intact and released with the knee in 0 and 90 degrees of flexion. The average pressure in the combined (knee 0 and 90 degrees flexed) sling intact group was ± mm Hg, compared with ± mm Hg in the combined (knee 0 and 90 degrees flexed) sling released group (P =.29). Although the pressure in the combined sling intact group did exceed the predefined critical value, this is due to the inclusion of the knee 0- degree flexed group. The average pressures experienced by the sciatic nerve in each group were as follows: sling intact knee 0 degree flexed, ± mm Hg; sling intact knee 90 degrees flexed, ± mm Hg; sling released knee 0 degree flexed, ± mm Hg; and sling released knee 90 degrees flexed, 55.3 ± mm Hg (Fig. 5). A significant difference in the pressure experienced by the sciatic nerve during hip reduction was present when isolating the variable of knee positioning (0 degree versus 90 degrees flexed; P =.002), with the higher pressure occurring when the knee was 0 degree flexed. Knee positioning was also found to be a significant variable in determining pressure Fig. 5. Graph showing comparisons during reduction with gentle retraction of the nerve. Horizontal line at 400 mm Hg shows predetermined pressure level at which nerve injury was considered possible. RED indicates hip reduction; SI, sling intact; SR, sling reduced; 0, knee at 0 degree of flexion; 90, knee at 90 degrees of flexion. Error bars show SD.

5 Total Hip Resurfacing Sciatic Nerve Injury Gay et al 1299 experienced by the sciatic nerve within both the sling intact (P =.05) and the sling released (P =.01) groups. When the condition of the sling (intact versus released) was isolated at both knee positions, it was not found to be a significant variable in determining pressure experienced by the sciatic nerve (P =.81 and P =.34). Thus, knee positioning and not the condition of the gluteal sling seems to play a significant role in pressures placed on the sciatic nerve during hip reduction. Discussion Since the approval for use in the United States in 2006, new-generation metal-on-metal total hip resurfacing has been an alternative to total hip arthroplasty. Recent early data have shown that the rate of nerve injury in modern total hip resurfacing is between 1.7% and 2.1% [6,7], compared with an incidence of approximately 1% in conventional total hip arthroplasty [2]. Modern total hip resurfacing has been indicated for use in a younger more active population, which means the dramatic impact of nerve injuries will be further increased as younger patients face the loss of function associated with nerve injury. Early US results suggest concerns about risk of nerve palsies in total hip resurfacing as compared with traditional hip arthroplasty. In a US study of 537 initial patients undergoing this procedure, Della Valle et al [6] reported a 1.7% incidence of postoperative nerve palsies at 1-year follow-up as compared with an incidence of 1% in traditional total hip arthroplasty [2]. An Australian study of 230 hips and mean 5-year follow-up found a 2.1% incidence of postoperative nerve palsies with modern total hip resurfacing [7]. Our findings suggest areas of potential risk of sciatic nerve injury in the standard total hip resurfacing technique done through a posterior approach. The highest pressures experienced by the sciatic nerve occurred during acetabular exposure, which is required for reaming and cup placement in total hip resurfacing. Of potential concern, in terms of the predefined pressure and duration representing nerve injury, is that reaming and inserting the acetabular cup clinically may require 15 minutes or more. The condition of the gluteal sling was the most significant variable in determining the pressure experienced by the sciatic nerve during acetabular exposure and preparation. Regardless of the position of the knee (0 or 90 degrees flexed), an intact gluteal sling resulted in pressures exceeding the critical value. Although fully extending the knee with the sling intact further increased the pressure placed on the sciatic nerve, it did not make a significant difference. Once the gluteal sling was released, the average pressures during acetabular exposure dropped below the critical value. Fully extending the knee raised the pressure placed on the nerve, but the difference was not significant. These findings suggest that releasing the gluteal sling during exposure may help to prevent postoperative sciatic nerve palsies. During positioning for femoral preparation, no variable resulted in a significant difference in the pressure placed on the sciatic nerve. Pressure on the nerve was significantly increased with the gluteal sling intact and the knee fully extended, but pressure did not exceed the predefined critical value. In contrast, position of the knee was a significant factor affecting pressure on the nerve during hip reduction. With the gluteal sling intact or released, a fully extended knee significantly increased the pressure placed on the sciatic nerve to raise the defined critical level during hip reduction, whereas pressure fell below the critical level when the knee was flexed 90 degrees. Although reduction of the hip with an extended knee alone probably does not cause a postoperative nerve palsy due to the transient nature of the pressure placed on the nerve, it may play a role if the nerve has been stressed in some manner earlier in the procedure. This potential for additional injury may be further exacerbated by multiple reductions if trailing is performed. These findings suggest the possible benefit of at least partially flexing the knee during hip reduction in this technique. There are several limitations within our study. There were large SDs in the pressure readings in several instances. Although the SDs shown in the graphs at times cross over the described critical pressure level, we believe the trends seen in the average pressures remain important. All procedures and all retractor positioning were done by the primary investigator, and blinding of this investigator with regard to pressure readings was maintained. We speculate that subtle differences in retractor positioning and varied patient anatomy may have played a role in the wide variations observed. In future studies with this model, the primary investigator could be notified of unusually high pressures allowing for a through examination of the relationship of the retractor to the nerve. We were also limited by a lack of information on the amount of pressure needed to cause a sciatic nerve injury, which required the use of presumed pressure levels and pressure duration based on an animal model [9]. Pressure level and duration associated with injury in humans may be higher or lower than that assumed in the current study, but the findings provide a useful relative comparison. Furthermore, cadaveric specimens may differ in their tissue compliance and nerve excursion with respect to living tissue. By using fresh (not frozen or embalmed) specimens, we feel that we have set up the best possible model to accurately reflect healthy living tissue. Finally, pressures during hip reduction were recorded with the native head and acetabulum with no implants inserted. The implants used in this technique clinically may change the

6 1300 The Journal of Arthroplasty Vol. 25 No. 8 December 2010 values obtained, but properly fitted implants should not change the findings substantially during hip reduction. These findings cannot be extrapolated to traditional total hip arthroplasty because the model was designed with an intact femoral head and neck replicating total hip resurfacing. In conclusion, the current findings suggest that releasing the gluteal sling during exposure with a posterior approach in total hip resurfacing may help to prevent postoperative sciatic nerve palsies. Consideration should also be given to partially flexing the knee during hip reduction in this procedure. Acknowledgment The authors thank Lyn Camire, ELS, of our department for editorial assistance. References 1. Schmalzried TP, Amstutz HC, Dorey FJ. Nerve palsy associated with total hip replacement. Risk factors and prognosis. J Bone Joint Surg Am 1991;73: Schmalzried TP, Noordin S, Amstutz HC. Update on nerve palsy associated with total hip replacement. Clin Orthop Relat Res 1997;344: Johanson NA, Pellicci PM, Tsairis P, et al. Nerve injury in total hip arthroplasty. Clin Orthop Relat Res 1983;179: Daniel J, Pynsent PB, McMinn DJ. Metal-on-metal resurfacing of the hip in patients under the age of 55 years with osteoarthritis. J Bone Joint Surg Br 2004;86: De Smet KA. Belgium experience with metal-on-metal surface arthroplasty. Orthop Clin North Am 2005; 36:203, ix. 6. Della Valle CJ, Nunley RM, Raterman SJ, et al. Initial American experience with hip resurfacing following FDA approval. Clin Orthop Relat Res 2009;467: Hing CB, Back DL, Bailey M, et al. The results of primary Birmingham hip resurfacings at a mean of five years. An independent prospective review of the first 230 hips. J Bone Joint Surg Br 2007;89: Hurd JL, Potter HG, Dua V, et al. Sciatic nerve palsy after primary total hip arthroplasty: a new perspective. J Arthroplasty 2006;21: Rydevik B, Lundborg G. Permeability of intraneural microvessels and perineurium following acute, graded experimental nerve compression. Scand J Plast Reconstr Surg 1977;11:179.