This article is more than 5 years old. Information may no longer be current.
Safe and Accurate: Learning the Direct Anterior Total Hip Arthroplasty
Abstract
Wear, instability, leg length, and muscle recovery are the major obstacles in total hip arthroplasty (THA). The direct anterior approach with fluoroscopic assistance has been proposed to address all four of these issues. The goal of this study was to assess the learning curve, safety, and accuracy of direct anterior THA. A retrospective review was completed on a single surgeon’s initial consecutive series of 300 THAs performed via a direct anterior approach. Cases were grouped based on chronologic order (1-100, 101-200, 201-300). Operative time, fluoroscopy time, estimated blood loss, radiographic leg length discrepancy (LLD), radiographic cup abduction angle, and complications were recorded. Data were analyzed using an analysis of variance. Postoperatively, all patients were managed without dislocation precautions. The average age for the cohort was 58.9 years, and the average BMI was 29. All groups were similar with respect to age and BMI. Direct anterior THA demonstrated significant reductions in operative and flouroscopy after the first 100 cases. Mean surgery time was 132.8 minutes for group 1, 109.9 for group 2, and 106.1 for group 3 (P <.001). Mean flouroscopy time was 32.1 seconds for group 1, 14.5 for group 2, and 14.5 for group 2 (P <.001). There was one dislocation in group 3 and three intraoperative calcar fractures in the first 100 cases. There were no infections. Direct anterior THA demonstrated a reduction in operative time and fluoroscopy time after the initial 100 cases. Calcar fracture did not occur after the first 62 cases. Cup abduction angle, dislocation rate, and LLD were excellent in all groups.
Despite the success of total hip arthroplasty (THA), wear, instability, leg length, and muscle recovery remain major obstacles after the procedure. Postoperative dislocation after the posterior approach and prolonged abductor recovery after anterolateral and direct lateral approaches have been documented.1 Inaccurate acetabular component placement via any approach has been shown to increase dislocation and, more importantly, increase wear.2 Increased bearing wear related to elevated cup abduction angles is becoming an even greater issue now that thin polyethylene liners and large hard-on-hard bearing surfaces are being used for younger patients undergoing THA. Although the large-diameter bearings in THA have reduced the incidence of dislocation, they have demonstrated the requirement for more accuracy in component placement to avoid elevated cup abduction angles. Elevated acetabular cup abduction angles have been correlated with an increased risk of highly crossed-linked polyethylene liner fracture and increased wear rates in metal-on-metal articulations.3 Along with an increase in the use of large-diameter heads and hard-on-hard bearings for THA, a simultaneous consumer and marketing-driven movement for less invasive surgery has emerged.
Unfortunately, early trends in less invasive THA focused on limiting surgical dissection, and therefore, limiting visualization that can lead to technical errors during surgery. Recent results of less invasive THA reveal that the patient benefit is only cosmetic, and some reports have noted inferior component alignment compared with conventional THA exposures.4-7
The direct anterior approach using fluoroscopic assistance has been proposed to address the issues of instability, component alignment, and muscle recovery. The purpose of this study was to assess the learning curve, safety, and accuracy of direct anterior THA performed in a single surgeon’s initial 300 consecutive cases.
Literature supporting the direct anterior approach for THA is limited. The history of the direct anterior approach for THA dates back to Judet’s report of acrylic hip arthroplasty in 1950.8,9 The first report of direct anterior THA in the United States was published by Keggi et al using a standard operating table.8 Matta et al and Bradley et al recently reported on the direct anterior approach using a specialized orthopedic extension table for THA.9,10 The only prospective study evaluating the direct anterior approach vs a miniposterior approach was presented by Nakata in 2007.11 Results revealed faster patient recovery of gait function and improved acetabular component alignment in the anterior vs miniposterior group.
Materials and Methods
We retrospectively reviewed a single surgeon’s initial unselected consecutive series of 300 THAs performed via a direct anterior approach. Cases were grouped based on chronologic order (case 1-100 = group 1, case 101-200 = group 2, and case 201-300 = group 3). Operative time, fluoroscopy time, estimated blood loss, and postoperative complications were recorded via retrospective review of hospital and clinic medical records. Intraoperative and postoperative adverse events, or complications, were defined as intraoperative fracture, dislocation, infection, neurologic injury, and revision surgery. Radiographic leg length discrepancy and radiographic cup abduction angle were measured on 1-, 4-, or 12-month postoperative pelvis radiographs.
Seven patients did not return to the clinic for postoperative follow-up. These cases were excluded from the calculation of postoperative complication rates including infection and dislocation. These cases were included for analysis of demographic, operative data, and intraoperative complications.
Standard descriptive statistics including frequency, mean, and range was calculated. A one-way analysis of variance with a Bonferroni correction was used to determine statistical differences among groups for parametric data including age, body mass index (BMI), follow-up, surgical time, fluoroscopy time, estimated blood loss, leg length discrepancy, and cup abduction angle.
Surgical Technique
The direct anterior approach uses the only true internervous/intermuscular plane to the hip via the short Smith-Peterson exposure. The approach allows preservation of the posterior soft tissues to prevent dislocation and rapid muscle recovery due to preservation of the abductor and deltoid mechanism of the hip. A specialized extension surgical table provides mobility and stability of the femur for the anterior approach through a single abbreviated incision and limited muscle dissection.
 Figure 1: Fluoroscopy of a cup placement. |
All surgeries were performed by a single surgeon (J.L.M.). Intraoperative cell salvage was used in all cases. An extension table was used for all anterior surgical approaches. The table allows for supine positioning of the patient, which permits improved component orientation and the use of fluoroscopy. Preoperative fluoroscopy was used to confirm pelvis positioning, leg length, and offset of the involved hip.
The limited direct anterior approach used the interval between the tensor fascia and the sartorius muscle. To prevent injury to the lateral femoral cutaneous nerve, the 10- to 12-cm skin incision and fascial incisions were performed lateral to the anterior superior iliac spine and lateral to the muscular interval. Hip exposure was obtained through an anterolateral capsulotomy with complete preservation of the abductor mechanism.
The femoral neck osteotomy was performed in situ before dislocation. A napkin ring segment of the femoral neck was removed to facilitate femoral head extraction with a power cork screw. Traction on the operative extremity was used to facilitate head extraction. The osteotomy level was initially referenced from the junction of the lateral femoral neck and greater trochanter. After exposure of the medial femoral calcar and lesser trochanter, the neck osteotomy was revised from a measurement referencing the lesser trochanter, if desired. Posterior capsular release was performed from the posterior femoral neck with the leg in external rotation. The femur was rotated back to within 30° of neutral for acetabular exposure.
The acetabular labrum was excised, and the reflected head of the rectus femoris was released from the anterior/superior acetabulum. The reflected head of the rectus femoris and the anterior capsule were reflected medially and tagged for later repair. Inferomedial capsulectomy was performed to identify the inferior border of the acetabulum. After complete acetabular exposure, reaming was performed under direct visualization. Fluoroscopy was used for final reaming and acetabular component insertion (Figure 1).
 Figure 2: Offset broach handle. |
Femoral exposure was performed via femoral extension, external rotation, and adduction. Elevation of the proximal femur was aided by use of a femoral hook placed outside the vastus origin around the subtrochanteric level of the femur. Adequate femoral exposure required anterior and lateral translation of the femur to clear the acetabular rim and the anterior superior iliac spine. The degree of femoral mobilization required depends on many factors including femoral neck length, capsular contracture, body size, and anatomic flare of the ilium. If additional mobilization of the femur was required, sequential release of the obturator internus and Gemelli was performed. The obturator externus, quadratus femoris, and gluteus maximus were always preserved.
Broaching of the femur was performed under direct visualization with offset broach handles (Figure 2). After insertion of the first broach, fluoroscopy was initially used to verify femoral broach alignment before proceeding with additional femoral preparation. With the desired broach in place, a trial reduction was performed by returning the extremity to neutral alignment. Traction followed by internal rotation facilitated reduction. Fluoroscopy was performed to evaluate component alignment, component size, leg length, and offset. With traction removed, soft-tissue tension was evaluated with femoral shuck test. Anterior stability was evaluated with extension and external rotation. Posterior stability was evaluated on the first 30 cases by removal of the traction boot from the leg spar, which then permits flexion and internal rotation. Posterior stability was not checked intraoperatively on the later 270 cases. Anterior capsular repair was performed, followed by a simple running closure of the tensor fascia. Skin closure was completed with a subcuticular absorbable monofilament suture.
Immediate weight bearing was permitted with both cemented and uncemented implants (exception: three calcar fractures). Because the posterior soft-tissue musculotendinous structures were either partially or completely left intact, unrestricted activity was permitted immediately after surgery. No patient was instructed on dislocation precautions. Patients were encouraged to resume ambulation and discontinue use of walking aids within the first 2 to 3 weeks. Warfarin was prescribed for 4 weeks postoperatively with a goal international normalized ratio of 2.0 to 3.0.
Results
Between September 2004 and May 2007, a single surgeon performed the first 300 consecutive primary total hip arthroplasties in 269 patients using a direct anterior approach. Of the 269 patients, there were 143 women and average patient age was 58.9 years (range, 15.6-90.2 years). The average BMI was 29.0 (range, 16.8-58.9). Of the 300 cases, 149 were right sided. The average follow-up was 13.8 months (range, 1-39.8 months) for the cohort. There were no statistical differences among groups with respect to age, sex, or BMI. There was a significant difference (P < .001) among groups in follow-up interval (Table 1).
Implants used
All surgeries were performed using an uncemented acetabular component (Table 2). Femoral components were uncemented in 286 and cemented in 14 cases. The decision to use cemented femoral fixation was based on intraoperative subjective femoral bone quality.
Bearing surfaces used included cobalt chrome on polyethylene, alumina on alumina, and cobalt chrome on cobalt chrome (Table 3).
Operative Time
Total surgery time including fluoroscopic imaging was recorded (Table 4). Of the 300 cases, the mean surgery time (incision to closure) was 117.8 (range, 74 - 210). There was a significant difference (P < .0001) between groups with respect to surgery time. Although the surgery time decreased over time, the largest change occurred between group 1 time period (mean, 132.8 minutes; range, 80 - 210 minutes) and group 2 (mean, 109.9 minutes; range, 78 - 194 minutes). The mean operative time for group 3 was 106.1 sec (range, 74 - 171 minutes).
Fluoroscopy Time
The mean fluoroscopy time was 20.3 (range, 0 - 192 sec) for the 300 cases (Table 4). There was a significant difference between groups for fluoroscopy time. The mean fluoroscopy time for group 1 was 32.1 sec (range, 0 - 192 sec), 14.5 sec (range, 0 - 102 sec) for group 2, and 14.5 sec (range, 0 - 162 sec) for group 3 (P < .001).
Leg Length Discrepancy
Of the 300 cases, the mean leg length discrepancy was 2.3 mm (0 - 15 mm) (Table 4). The mean leg length discrepancy significantly (P < .001) improved over time. The mean LLD for group I was 3.5 mm (range 0 - 15 mm), 2.3mm (range, 0 - 13 mm) for group 2, and 1.1 mm (range, 0 - 8 mm) for group 3.
Cup Abduction Angle
The mean cup abduction angle for the cohort was 43.0º (26º - 53º) (Table 4) There were significant differences (P < .001) between groups with respect to acetabular component positioning. The mean cup abduction angle was 41.8º (range, 30º = 53º) for group I, 42.8º (range, 26º = 53º) for group 2, and 44.4º (range, 32.0º = 52.0º) for group 3.
Complications
Dislocation
There was 1 posterior dislocation (group 3, case 212). The dislocation occurred 9 months postoperatively while the patient was putting on shoes with the involved leg elevated on stairs. The dislocation was treated by closed reduction and did not recur. This case revealed normal cup abduction alignment of 42º and a 32-mm alumina articulation.
Calcar Fracture
Intraoperative calcar fracture occurred in 3 hips (all in group 1). All 3 calcar fractures were recognized intraoperatively and treated with proximal cerclage cable placement through the same incision, reinsertion of the uncemented tapered femoral stem, and protected weight bearing for 6 weeks. All three fractures proceeded to clinical and radiographic union, and no femoral component was revised or demonstrated subsidence.
Neurologic Injury
No patients demonstrated a neurologic motor deficit postoperatively. Varying degrees of anesthesia, or numbness, lateral to the anterior incision were seen. No cases of meralgia paresthetica were seen.
Infection
There was 1 deep infection that presented at 4 months postoperatively. This patient was treated with urgent irrigation and debridement, retention of components, and 6 weeks of intravenous antibiotic therapy. At 2 years postoperatively, this patient presented with elevated inflammatory serum markers and has since undergone a successful two-stage revision.
Wound Complications
Three patients required operative irrigation and debridement of superficial wound seromas during the immediate postoperative period. Deep cultures on all 3 cases were negative for microbial growth, and all 3 patients have proceeded to uneventful healing of their incisions. The BMI for all 3 patients was over 40.
Revision Surgeries
Four revision surgeries were performed. One for deep infection at 4 months postoperatively (described above), 1 femoral revision for fibrous fixation and thigh pain, 1 for thigh pain with an undersized femoral stem suspected to be loose but was found to be solidly ingrown at the time of revision, and 1 for squeaking after an alumina-on-alumina bearing that was revised to cobalt chrome on polyethylene with retention of well-fixed and well-positioned femoral and acetabular components.
Discussion
Our results in this unselected consecutive initial series demonstrate significant improvements in efficiency during the initial 300 cases. The direct anterior THA demonstrated reduced operative and fluoroscopy times after 100 cases. The reduction in operative and fluoroscopy times after 200 cases was not significant.
Our results demonstrate that cup abduction angle, leg length, and dislocation rate were satisfactory in all groups and did not show improvement with increasing surgical experience.
Calcar fracture was common during the first 62 cases (5%). It should be noted that multiple uncemented stem designs were implanted during this part of the series and this, along with a new femoral exposure, may be partially responsible for the increased incidence. After 62 cases and limitation of uncemented stems to two designs, no additional calcar fractures were seen. Importantly, all 3 calcar fractures were identified intraoperatively and treated successfully with uncompromised WOMAC scores at 4 months follow-up. This is in contrast to minimally invasive hip surgery via a 2-incision approach that does not allow direct visualization of the femur and intraoperative recognition of calcar fractures that have been reported to occur in approximately 1% to 4% of uncemented tapered femoral stems, even through a conventional surgical approach.12
Wound-healing issues appear to be partially related to BMI. We encountered 3 cases of wound complication/seroma that required surgical treatment. Each of these patients was morbidly obese based on BMI calculations. It should be noted that the dermis covering the anterior portion of the hip is significantly thinner than the lateral and posterior dermis and is less resistant to soft-tissue trauma. We recognized this during our experience and have made two changes to our surgical technique based on these findings. Sharp dissection using a scalpel is used until the tensor fascia is reached during the initial exposure. We believe this prevents iatrogenic subcutaneous-tissue injury associated with electrocautery. Second, we close all incisions in morbidly obese patients using vertical nylon mattress suture and keep the sutures in place for 10 to 14 days.
Conclusion
It appears that medial femoral calcar fracture was the only complication that occurred at an increased frequency during the initial portion of the series and did not occur after the first 62 cases. Operative and fluoroscopy times were reduced significantly throughout the first 100 cases.Based on these results, we feel that the learning curve regarding safety for the direct anterior approach is related to femoral exposure and broaching with modified instruments and can be shortened by using a single stem and broach design conducive to the anterior approach. Specifically, an offset broach handle and a shorter stem implanted through a broach-only technique with a less prominent lateral shoulder appear to be the optimal design. The learning curve for time appears to continue through the first 100 cases with significant reductions in time and fluoroscopy use thereafter. The learning curve for accuracy and patient outcomes appears to be flat and is most likely attributable to the use of intraoperative fluoroscopy and a true intermuscular surgical approach.
References
- Masonis J, Bourne R. Surgical approach, abductor function, and total hip arthroplasty dislocation. Clin Orthop. 2002;405:46-53.
- Patil S, Bergula A, Chen P, et al. Polyethylene wear and acetabular component orientation. J Bone Joint Surg Am. 2003;85-A (Suppl 4):56-63.
- Tower SS, Currier JH, Currier BH, et al. Rim cracking of the cross-linked longevity polyethylene acetabular liner after total hip arthroplasty. J Bone Joint Surg Am. 2007;89(10):2212-7.
- Woolson S, Mow C, Syquia J, et al. Comparison of primary hip replacements performed with a standard incision or mini-incision. J Bone Joint Surg Am. 2004; 86-A 1353-8.
- Wright, Sculco et al. Mini-incision for total hip arthroplasty: prospective controlled investigation with 5 year follow-up evaluation. J Arthroplasty. 2004;19: 538-45.
- DeBeer J, Petruccelli D, Zalzal P, et al. Single incision, minimally invasive total hip arthroplasty; length doesn’t matter. J Arthroplasty. 2004; 19(8):945-50.
- Kim Y. Comparison of primary total hip arthroplasties performed with a minimally invasive technique or a standard technique. A prospective and randomized study. J Arthroplasty. 2006; 21:1092-8.
- Judet J. Protheses en resine acrylic. Mem Acad de Chir. 1947;73:561.
- Judet J. The use of an artificial femoral head for arthroplasty of the hip joint. J Bone Joint Surg [Br]. 1950;32-B:166-73.
- Light T, Keggi, K. Anterior approach to hip arthroplasty. Clin Orthop. 1980;152:255-60.
- Matta J, Shahrdar C, Ferguson T. Single incision anterior approach for total hip arthroplasty on an orthopaedic table. Clin Orthop. 2005;441:115-24.
- Nakata K, Masataka N, Koji Y, et al. A prospective comparative study of the direct anterior with mini-posterior approach in MIS-THA. AAOS Meeting. February 15, 2007. San Diego.
- Berend K, Lombardi A Jr, Mallory T, et al. Cerclage wires or cables for the management of intraoperative fracture associated with a cementless, tapered femoral prosthesis: results at 2 to 16 years. J Arthroplasty. 2004;19(7 Suppl 2):17-21.
Authors
Dr Masonis is from the OrthoCarolina Hip & Knee Center; and Caryn Thompson, BS, CCRC, and Susan Odum, M.Ed, CCRC, are from the OrthoCarolina Research Institute, Charlotte, North Carolina.
Dr Masonis is a consultant for and received grant support from Smith & Nephew Orthopedics. Caryn Thompson and Susan Odum have no relevant financial relationships to disclose.
The investigators thank Amanda Phillips and Anne Dennos of the OrthoCarolina Research Institute for their assistance with this project and preparation of the manuscript.
Correspondence should be addressed to Dr Masonis, OrthoCarolina Hip & Knee Center, 1025 Morehead Medical Drive, Charlotte, NC 28204.