Secondary Logo

Journal Logo

A Clinical Comparison of the Anterolateral and Posterolateral Approaches to the Hip

Ritter, Merrill, A.; Harty, Leesa, D.; Keating, Michael, E.; Faris, Philip, M.; Meding, John, B.

Clinical Orthopaedics and Related Research: April 2001 - Volume 385 - Issue - p 95-99

Patients who had anterolateral and posterolateral approaches in total hip replacement surgery were compared clinically for limp, dislocation, hospital stay, and discharge disposition. The only statistical difference was that the posterior approach had a statistically higher dislocation rate. Although the number of patients with limp was higher in the anterolateral group, the difference was not statistically different.

From the Center for Hip and Knee Surgery, A Division of Orthopaedics Indianapolis, Inc., St Francis Hospital—Mooresville, Mooresville, IN.

Reprint requests to Merrill A. Ritter, MD, 1199 Hadley Road, Mooresville, IN 46158.

Received: September 17, 1999.

Revised: April 3, 2000; July 17, 2000.

Accepted: August 15, 2000.

Two basic approaches to primary total hip replacements have evolved since Charnley’s use of the direct lateral approach via trochanteric osteotomy. Because of trochanteric nonunion and its resulting dislocation and disability rates, trochanteric osteotomy generally is reserved for revision arthroplasty. The anterolateral and posterior approaches are now the more commonly used approaches. Each approach has been evaluated and the advantages and disadvantages of each elucidated. 1–3,7–9,12,13,15 Both approaches appear to allow decreased blood loss 8,13 and fewer hematomas 13 when compared with the transtrochanteric approach. The anterolateral approaches generally are thought to have lower dislocation rates 2,10,12 and allow excellent acetabular exposure although limping is increased (at least temporarily). 1,2 The posterior approach may allow lower blood loss than the anterolateral approach 13 and maintenance of abductor strength 3 but it generally results in a higher dislocation rate. 7,9,15

Although the posterior approach has remained essentially unchanged, the anterolateral approach has been modified by several surgeons to decrease gluteus medius disruption and hopefully to decrease postoperative abductor muscle dysfunction and resultant limp. 1,2,4,9,12 However, no studies from the past decade have been found that compared abductor muscle dysfunction between the posterior approach and a modified anterolateral approach. Although Baker and Bitounis, 1 Frndak et al, 2 and Gore et al 3 imply weakening of the abductor muscles as a result of the anterolateral approach, only the study by Gore et al 3 showed a statistically significant increase in weakness of the abductors in the anterolateral approach.

As in the case of abductor muscle function, the instance of dislocation resulting from the anterolateral and posterior approaches has not been compared in the same study in the past decade. Mallory et al 9 compiled tables of dislocation rates by approach as found in various studies. The tables appear to indicate a generally higher instance of dislocation in the posterior approach, but they merely indicate a possibility that the posterior approach leads to more dislocations. A study including data from the same institution and statistical analysis would be beneficial.

The authors were interested in the early postoperative period of anterolateral and posterior approach total hip replacements regarding length of hospital stay and discharge disposition, but no articles that dealt with these issues were found.

The purpose of the current study was to address questions about dislocation and limping in patients and to determine whether early rehabilitation time differed between the two approaches based on hospital stay and discharge disposition. The study compared the clinical results of total hip replacements performed at the same institution by four experienced hip surgeons using the anterolateral and posterior approaches. Approval was obtained from the hospital Institutional Review Board for this retrospective review.

Back to Top | Article Outline


Three hundred twenty consecutive primary total hip replacements were done on 302 patients in 1997. Of these hips, 130 total hips arthroplasties in 121 patients were done by one author (MAR) using the modified anterolateral approach as described by Mulliken et al. 12 In this group, there were 52 (43%) men and 69 (57%) women. The average age was 68 years (range, 38–89 years). The preoperative diagnosis was osteoarthritis in 120 hips (92.3%) (Table 1). Cemented and uncemented femoral and acetabular components were used. A cemented, all-polyethylene acetabular component was used in 94 hips (72.3%) and a cemented femoral component was used in 90 hips (69.2%). All femoral heads were 28 mm in diameter (Table 2). Differing neck lengths were used to obtain proper soft tissue tension.





The posterolateral approach was used by the other three surgeons (EMK, PMF, JBM) in 190 hips on 181 patients. There were 82 (45.3%) men and 99 (54.7%) women. The average age was 67 years (range, 32–89 years). Osteoarthritis was the primary diagnosis in 169 (88.9%) patients (Table 1). A cemented, all-polyethylene acetabular component was used in 125 hips (65.8%) and a cemented femoral component was used in 152 hips (80%) (Table 2). Twenty-eight millimeter heads were used in each procedure and neck lengths were adjusted for appropriate soft tissue tension. The short external rotators and posterior capsule were not repaired in this series.

Patients were evaluated for limp and dislocation using the Harris hip score. 6 All patients who were seen in the clinic for a 1-year followup and all patients who could be contacted by telephone or letter were included in the study. Certain objective findings in the Harris hip score such as range of motion (ROM) and absence of deformity cannot be evaluated in a telephone interview; however, these scores constitute a minimal contribution to the total score. Patients with telephone interviews were questioned whether any change in their hips had occurred since the last clinic visit. As the patients stated that no changes had occurred, the patients’ scores from the most recent clinic visit were used in the ROM and absence of deformity categories.

Limp was not categorized as to type. All limp, as observed by the evaluator or perceived by the patient, was categorized as a positive response. Likewise, dislocations were not categorized as to etiology or recurrence. Any dislocation or dislocations were classified as one positive response for that patient.

Three other dependent variables also were compared between the groups and included length of stay, infection, and discharge disposition.

T tests were used to compare average hip scores and length of stay. A Wilcoxon rank sum test was used to compare average pain scores. Fisher’s exact test was used to compare gait scores and presence or absence of limp. Chi square tests were used to compare frequency of types of discharge disposition, and the frequency of dislocation.

Back to Top | Article Outline


Demographic Data

Patients with 96 of the anterolateral approach to the hip and 190 of the posterior approach were evaluated in the clinic at the patients’ 1-year followup. Patients with 26 of the anterolateral approach to the hip and 38 with the posterior approach were evaluated by letter or telephone call. Eight patients with eight anterolateral total hip replacements either were deceased (four patients) or lost to followup (four patients). Three patients with three posterior total hip replacements were deceased and three patients with three posterior total hip replacements hips were lost to followup. Patients who were lost to followup were included in the length of hospital stay and disposition analysis, but were excluded from the remainder of the comparison. Average followup for the anterolateral approach was 14.8 months (range, 9.2–28.2 months), and for the posterior approach the average followup was 14.7 months (range, 10–29 months). Discharge data are provided in Table 3.



Back to Top | Article Outline


The average Harris hip score for the anterolateral approach was 92.63 ± 12.20 (range, 35–100 points) and the average pain score was 41.66 ± 6.07. For the posterior approach, the average Harris hip score was 92.03 ± 10.15 (range, 25–100 points), and the average pain score was 41.91 ± 4.98. Limp of any severity was seen in 35 (28.69%) of the patients with the anterolateral approach, nine (7.38%) of whom had a moderate or severe limp (Table 4). Limp of any severity was seen in 31 patients with the posterior approach (16.84%) and moderate or severe limp was seen in seven patients (3.8%) (Table 4).



Back to Top | Article Outline

Length of Stay and Discharge Disposition

The average length of stay for patients with the anterolateral approach was 5.25 days ± 11.39 (range, 3–14 days). The average length of hospital stay for patients with the posterior approach was 5.19 days ± 0.95 (range, 2–10 days). Eighty-one (62.31%) patients who had the anterolateral approach were discharged to home or self-care and seven (5.38%) were discharged to rehabilitation facilities (Table 3). One hundred thirty-two patients (69.47%) who had the posterior approach were discharged home or to self-care and 12 patients (6.32%) were transferred to rehabilitation facilities. One patient with the anterolateral approach and one patient with the posterior approach had an infection.

Back to Top | Article Outline


No patient with the anterolateral approach experienced dislocation. There were eight (4.21%) hips with the posterior approach in which dislocation occurred. Five of the dislocations were anterior and three were posterior.

Back to Top | Article Outline


Of the many factors that have been used to compare the anterolateral and posterolateral approaches to the hip (blood loss, time of operation, ectopic bone formation, length of stay, neurologic injury, limp, and dislocation), the most persistent differences are perceived to be postoperative limp and postoperative instability (dislocation). Articles by Roberts et al, 13 Turner, 14 Vicar and Coleman, 15 and Woo and Morrey, 16 have detailed a higher dislocation rate with the posterolateral approach as compared with the anterolateral approach as shown by Mallory et al 9 and Mulliken et al. 12

The current findings support these observations with eight dislocations occurring in patients in the posterolateral group as opposed to no dislocations in the anterolateral group (p = 0.018). McCollum and Gray 10 suggested that this increased dislocation rate might be attributable to inadequate acetabular exposure and consequent malposition of the acetabular component. Five of the eight dislocations in patients in the study were anterior. This suggests two possible etiologies, one being excessive anteversion of the acetabular component perhaps as overcompensation for the tendency to retrovert the component and for a weakened or absent posterior capsule or both. The second possible etiology is the desire to remove anterior impingement sources, leading to overly aggressive removal of anterior osteophytes and anterior capsule, resulting in secondary weakening of the anterior restraints to instability. In the anterolateral approach, the posterior capsule and osteophytes are left intact.

All four surgeons have many years experience with the approach of their choice. Acetabular exposure is thought to be adequate in both approaches and orientation of the acetabular component was done using standard landmarks; however, special radiographic techniques to measure anteversion were not done and consequently acetabular position is not compared in this study.

Postoperative limp has been cited as the biggest objection to the anterolateral approach. Hardinge 4 described a modification of the direct lateral approach to the hip reported by McFarland and Osborne. 11 This involved splitting the gluteus medius continuous with the anterior aspect of the vastus lateralis. Criticism of this approach included prolonged limp and partial denervation of the gluteus medius. 1,5 The anterolateral approach used in this study was a slight modification of that described by Mulliken et al 12 in which the gluteus medius split proceeded anteriorly from the superior tip of the greater trochanter, differing from that described by Hardinge. 4 With these modifications, limp seems to be less of a problem. The current study showed no significant difference in limp although there was a greater percentage of patients in the anterolateral group who limped (21.3% versus 13%). Greater patient numbers would be required to establish a statistical difference. No other significant differences were found between the two groups in Harris hip scores, length of hospital stay, or discharge disposition.

Based on the current study, there appears to be a trade off between the two approaches as related to dislocation versus limp. However, with the newer modifications of the anterolateral approach, it appears that postoperative limping is less of a problem and that the incidence of dislocations as a result of using the posterior approach remains high.

Back to Top | Article Outline


1. Baker AS, Bitounis VC: Abductor function after total hip arthroplasty: An electromyographical and clinical review. J Bone Joint Surg 71B:47–50, 1989.
2. Frndak PA, Mallory TH, Lombardi AV: Translateral surgical approach to the hip: The abductor muscle “split”. Clin Orthop 295:135–141, 1993.
3. Gore DR, Murray MP, Sepic SB, et al: Anterolateral compared to posterior approach in total hip arthroplasty: Differences in component positioning, hip strength, and hip motion. Clin Orthop 165:180–187, 1982.
4. Hardinge K: The direct lateral approach to the hip. J Bone Joint Surg 64B:17–19, 1982.
5. Hardy AE, Synek V: Hip abductor function after the Hardinge approach: Brief report. J Bone Joint Surg 70B:673, 1988.
6. Harris WH: Traumatic arthritis of the hip after dislocation and acetabular fractures: Treatment by mold arthroplasty. J Bone Joint Surg 51A:737–755, 1969.
7. Hedlundh U, Ahnfelt L, Hybbinette CH, et al: Surgical experience related to dislocation after total hip arthroplasty. J Bone Joint Surg 78B:206–209, 1996.
8. Horwitz BR, Rockowitz NL, Goll SR, et al: A prospective randomized comparison of two surgical approaches to total hip arthroplasty. Clin Orthop 291:154–163, 1993.
9. Mallory TH, Lombardi AV, Fada RA, et al: Dislocation after total hip arthroplasty using the anterolateral abductor split approach. Clin Orthop 358:166–172, 1999.
10. McCollum DE, Gray WJ: Dislocation after total hip arthroplasty: Causes and prevention. Clin Orthop 261:159–170, 1990.
11. McFarland B, Osborne G: Approach to the hip: A suggested improvement. J Bone Joint Surg 36B:367–370, 1954.
12. Mulliken BD, Rorabeck CH, Bourne RB, et al: A modified direct lateral approach in total hip arthroplasty: A comprehensive review. J Arthroplasty 13:737–747, 1998.
13. Roberts JM, Fu FH, McClain EJ, et al: A comparison of the posterolateral and anterolateral approaches to total hip arthroplasty. Clin Orthop 187:205–210, 1984.
14. Turner RS: Postoperative total hip prosthetic femoral head dislocations: Incidence, etiologic factors, and management. Clin Orthop 301:196–204, 1984.
15. Vicar AJ, Coleman CR: A comparison of the anterolateral, transtrochanteric, and posterior surgical approaches in primary total hip arthroplasty. Clin Orthip 188:152–169, 1984.
16. Woo RYG, Morrey BF: Dislocations after total hip arthroplasty. J Bone Joint Surg 64A:1295–1306, 1982.
© 2001 Lippincott Williams & Wilkins, Inc.