The advantages of minimally invasive surgical approaches for total hip arthroplasty are reported to include reduced blood loss, less pain, and shorter hospital stays, which combine to afford a faster recovery1-4. However, other studies have failed to show any significant advantage over the use of standard surgical approaches (Fig. 1)5-8. Reported disadvantages of minimally invasive techniques are the substantial learning curve required and an increased risk of early complications3,9-12. In addition, the long-term outcomes of minimally invasive procedures in terms of implant fixation and longevity remain unproven.
There are three basic categories of so-called minimally invasive total hip arthroplasty approaches: an abbreviated incision (small incision), modifications of standard approaches with smaller incisions and less soft-tissue dissection (less invasive), and novel approaches that reportedly do not cut muscle (minimally invasive). We have had a broad experience, over the past seven years, with a less invasive modification of the direct lateral approach (a modified Hardinge approach) (Fig. 2). We previously reported less blood loss and a shorter hospital stay in association with this approach13. Others have argued that the less invasive direct lateral approach does not provide an advantage over the traditional approach5. Importantly, the soft-tissue dissection still requires removing and repairing the abductor musculature.
Several recent reports have described various techniques of minimally invasive total hip arthroplasty performed through the Smith-Petersen interval with use of a single anterior incision4,14-16. Early reports on the use of this anterior approach in hip arthroplasty include those by Judet and Judet17 and Light and Keggi18, with the latter authors often incorporating multiple accessory incisions necessitated by older-style instruments. The anterior interval is both intermuscular and internervous, so the anterior approach has the potential advantage of requiring little or no muscle dissection and is a true minimally invasive alternative. These anterior-based approaches, like most other less invasive or minimally invasive approaches, now are facilitated by specialized instrumentation. The anterior supine intermuscular technique is one such approach (Fig. 3) and is performed with use of Microplasty ASI instrumentation (Biomet, Warsaw, Indiana).
Matta et al. described the use of specialized or modified operating fracture tables, similar to those used for pelvic fracture surgery14. Potential disadvantages of using a fracture table include the cost of the table, difficulty in checking the range of motion and stability, and equipment-related fractures. In two reports, the authors noted ankle fractures related to the use of these specialized tables, which require the foot to be fixed in a holder14,15. Alternative techniques include the use of a standard radiolucent operating table with extreme hyperextension and the use of a table-mounted femoral elevator with the lower limbs draped free.
The purpose of this study was to compare the early outcomes of primary total hip arthroplasties performed through an anterior supine intermuscular approach with the results of primary total hip arthroplasties done through a less invasive direct lateral approach; all of the procedures were carried out with the same cementless, shortened, tapered femoral component. We examined perioperative complications and early outcomes to determine whether the anterior supine intermuscular approach has an advantage over the less invasive direct lateral approach in terms of early recovery. A detailed review of our operative approach is also presented.
The records on all total hip arthroplasties performed between January 2006 and August 2008 with a Taperloc Microplasty stem (Biomet) by two surgeons (A.V.L. Jr. and K.R.B.) were examined. The first total hip arthroplasty through the anterior supine intermuscular approach was performed by a surgeon (K.R.B.) at our institution in February 2007. To shorten the learning curve, the surgeon participated in three training sessions with a total of five cadavers and carefully selected the first patients to decrease the difficulty of the procedure. Three categories of approaches—anterior supine intermuscular, less invasive direct lateral, and standard direct lateral—were utilized between January 2006 and August 2008. All patients were prospectively entered into our institutional review board-approved clinical database and electronic medical record (DocuMed, Ann Arbor, Michigan). The database was retrospectively reviewed for patient demographics, surgical and hospital data, and follow-up data. The patient demographics that were evaluated included age, sex, diagnosis, height, weight, and body mass index. Surgical data included operative duration (from incision to closure), estimated blood loss, and intraoperative complications. Hospital data included allogeneic blood transfusions and the length of the hospital stay. Follow-up data obtained at the six-week examination and any subsequent office visits included the standard variables identified with the Harris hip score19: pain, walking ability, use of assistive devices, and limping. Both preoperative and postoperative scores on a lower-extremity activity scale20 were also calculated. Data were compared between the group treated with the anterior supine intermuscular approach and that treated with the less invasive direct lateral approach.
The standard direct lateral approach that was utilized was previously described by Frndak et al.21 and is performed with the patient in the lateral decubitus position (Fig. 1). The less invasive direct lateral approach involves an abbreviated skin and fascial incision with limited abductor muscle dissection as compared with the dissection used in the standard approach (Fig. 2). This less invasive direct lateral approach is performed with the patient in the lateral decubitus position and has also been previously described13.
The anterior supine intermuscular approach (Fig. 3) is performed with the patient positioned supine on a standard radiolucent operating table with an extender at its foot. Fluoroscopy is used in every case. The patient is positioned with the pubic symphysis at the table break to allow subsequent positioning during femoral preparation and implant insertion (Fig. 4). Both lower limbs are prepared and draped for positioning during surgery. The proximal end of an 8 to 10-cm-long skin incision is placed approximately two fingerbreadths distal and two fingerbreadths lateral to the anterior superior iliac spine (Fig. 5). The incision is placed in this lateral position (as opposed to the more medial position used for the anterior incision of two-incision techniques) to avoid the lateral femoral cutaneous nerve. The correct position of the planned incision is confirmed with fluoroscopy (Fig. 6). The fascia overlying the tensor fasciae latae muscle is split throughout the length of the incision, and the muscle is bluntly dissected away from the interval with the sartorius muscle, with care taken to preserve the fascia between the two to protect the sartorius muscle belly. The deep muscle fascia of the tensor fasciae latae is then split, exposing the lateral circumflex vessels (Fig. 7). These vessels are ligated, cauterized, and transected. With blunt dissection, retractors are placed over and under the femoral neck. The inferior retractor lies under the rectus femoris muscle and over the vastus lateralis. A sharp, pointed retractor is carefully used to peel the rectus femoris off the anterior aspect of the capsule and then placed over the anterior rim of the acetabulum.
The exposed anterior aspect of the capsule is then excised, exposing the femoral neck. Meticulous capsular resection at this step makes subsequent steps of acetabular preparation easier. The inferior and superior femoral neck retractors are replaced intracapsularly, and the femoral neck is resected on the basis of preoperative templating and the resection is confirmed with fluoroscopy. A subcapital resection and final neck resection are performed to create a "napkin ring" slice of femoral neck. A threaded Steinmann pin is used to sequentially remove the napkin ring and the femoral head (Fig. 8). A sharp retractor is placed superiorly (at the three o'clock position for a left hip), a double-pronged posterior retractor is placed on the ischium (at approximately seven o'clock for a left hip), and an anterior retractor is placed if necessary (Fig. 9). The labrum, osteophytes, and any capsule that lies in the way of acetabular preparation are removed. Sequential reaming is performed, with use of fluoroscopic guidance as necessary (Fig. 10). The acetabular component is then placed, and its position is confirmed with fluoroscopy. Screws may be placed and a liner may be inserted, or a solid metal-on-metal articulation may be used.
Femoral preparation and implant insertion require specialized positioning. The anesthesiologist must jackknife the table by dropping its foot and placing the bed into a steep Trendelenburg position (Fig. 11). The patient's contralateral foot (on the side not being operated on) is placed on a padded Mayo stand. A table-mounted femoral elevator retractor (Omni-Tract Surgical, St. Paul, Minnesota) is placed on the bed, and the traction hook is placed around the proximal part of the femur, proximal to the conjoined tendon of the gluteus maximus. The femur is retracted to tension the capsule. A retractor is placed gently, directly proximal to the greater trochanter. This creates a fold of the superior aspect of the capsule that can be excised easily. The trochanteric retractor is then replaced to elevate the proximal part of the femur. The superior aspect of the capsule is dissected from the trochanter from anterior to posterior. With increasing gentle retraction applied through the table-mounted hook, the femur is elevated. Simultaneously, the operatively treated limb is externally rotated and adducted underneath the contralateral limb in a lazy figure-of-four position. Extreme knee flexion tightens the rectus femoris, making exposure more difficult. If the inferior aspect of the capsule has not been excised, it is then dissected from the femoral neck.
Femoral preparation is performed next. The use of a broach-only stem design is recommended as direct straight reaming of the femur is difficult in most cases and specialized offset broaches and broach handles make femoral preparation substantially easier. The final broach is left in place, a trial femoral neck and head is inserted, and the hip is relocated by releasing the traction of the table-mounted femoral elevator, internally rotating, abducting, and extending the limb. Fluoroscopic images are obtained to confirm femoral implant positioning, offset, and neck length as well as lower-limb length. The surgeon can directly measure lower-limb length with the patient in the supine position by comparing the medial malleoli and confirming the assessment with fluoroscopy (Fig. 12).
The hip is dislocated, and the final implant is inserted (Fig. 13). A trial reduction can then be performed to assess lower-limb lengths, and final fluoroscopic images are obtained. The bed is returned to the flat position, the wound is irrigated, a deep drain is placed, and the wound is closed in layers. A running suture for the superficial tensor fascia, interrupted subcutaneous sutures, and running subcuticular sutures are used. The incision is then closed with skin glue (DERMABOND; Ethicon, Somerville, New Jersey).
Standardized hospitalization and rehabilitation protocols were used in all cases, as previously reported22. The perioperative program includes the use of preemptive pain and nausea medications, intrathecal spinal anesthesia, long-acting narcotic analgesics, and a periarticular soft-tissue injection cocktail (60 mL of 0.5% ropivacaine, 30 mg of ketorolac, and 0.5 mL of 1:1000 epinephrine). This program as well as our multimodal approach to the prevention of venous thromboembolism has been previously reported23. Patients walk with a walker on the day of the surgery and are discharged from the hospital when they have accomplished physical therapy goals. They are instructed to use a walker for two weeks and then progress to the use of a cane or no assistive devices when they are able. Routine follow-up physical and radiographic examinations are performed at six weeks and annually thereafter.
Source of Funding
No external funding was received in support of this work.
There were 655 total hip arthroplasties performed with use of the Taperloc Microplasty stem (Biomet) in 605 patients during the study period: twenty-five (4%) of the 655 were done through a standard direct lateral approach, 372 (57%) were done through a less invasive direct lateral approach, and 258 (39%) were done through an anterior supine intermuscular approach. The sex distribution and average age were similar between the patients treated with the anterior supine intermuscular approach and those managed with the less invasive direct lateral approach; 44% (264) of the 605 patients were male, and the average age was sixty-three years in each group. The disease profiles and Charnley classifications were also similar between the two major treatment groups, with osteoarthritis being the underlying diagnosis in the majority of cases (83% of those treated with the anterior supine intermuscular approach and 80% of those managed with the less invasive direct lateral approach) and the distribution of the Charnley classes being A, B, and C in 43%, 36%, and 21%, respectively, of those treated with the anterior supine intermuscular approach and 43%, 38%, and 19% of those managed with the less invasive direct lateral approach. While the operative times for the anterior supine intermuscular approach were initially longer than those for the less invasive direct lateral approach, this difference disappeared over the course of the study period (sixty-nine minutes for the anterior supine intermuscular approach compared with sixty-eight minutes for the less invasive direct lateral approach; p = 0.7) (Fig. 14). The average blood loss during surgery, while low with either approach, was significantly greater for the anterior supine intermuscular approach (155 mL) than it was for the less invasive direct lateral approach (138 mL) (p = 0.006), but the overall risk of transfusion was no different between the two groups (transfusion was given after 4% [eleven] of the 258 operations done with the anterior supine intermuscular approach and 3% [ten] of the 372 done with the less invasive direct lateral approach; p = 0.3).
Despite a clear bias toward implementing the anterior supine intermuscular approach in easier cases during our early experience, there was no difference in the average height or weight of the patients treated with the anterior supine intermuscular approach and those of the patients managed with the less invasive direct lateral approach; however, the average body mass index of the former was lower than that of the latter (28.9 compared with 30.4 kg/m2; p = 0.004). The average preoperative Harris hip score was 50 points in each group, and the preoperative scores on the lower-extremity activity scale were similar (9.7 points for the patients treated with the anterior supine intermuscular approach compared with 9.1 points for those managed with the less invasive direct lateral approach; p = 0.1).
The average length of the hospital stay (Fig. 15) was similarly short for both groups (1.8 days for the patients treated with the anterior supine intermuscular approach compared with 2.0 days for those managed with the less invasive direct lateral approach; p = 0.1). However, the percentage of patients discharged to home rather than to a rehabilitation, skilled nursing, or extended-care facility was significantly higher in the group treated with the anterior supine intermuscular approach (87%; 207 of 239 [information not available for nineteen patients]) than it was in the group managed with the less invasive direct lateral approach (79%; 276 of 349 [information not available for twenty-three patients]) (p = 0.04).
While the average preoperative Harris hip scores were the same in the two treatment groups, the average six-week Harris hip score was significantly higher for the patients treated with the anterior supine intermuscular approach than it was for those managed with the less invasive direct lateral approach (80 compared with 75 points; p = 0.0000) (Fig. 16). Despite no significant difference in the preoperative scores on the lower-extremity activity scale, the average score on this scale at six weeks was significantly higher, demonstrating a faster return to function and daily activities, in the group treated with the anterior supine intermuscular approach than it was in the other treatment group (8.6 compared with 8.0 points; p = 0.03) (Fig. 17).
There were four intraoperative complications in the group treated with the anterior supine intermuscular approach, and all were identified and corrected during the procedure (Table I). There were two intraoperative proximal femoral perforations, which occurred during preparation for insertion of the stem; these were noted intraoperatively and did not alter the surgical procedure. In addition, one acetabular component dislodged, requiring acute revision to a modular device with screws, and one intraoperative pelvic fracture occurred during cup insertion, requiring femoral head autograft and the use of a porous metal revision acetabular component. There was one intraoperative complication in the group treated with the less invasive direct lateral approach; this was a nondisplaced small femoral shaft fracture, which was addressed with cerclage cable fixation at the time of surgery.
There were six complications requiring a reoperation in the group treated with the anterior supine intermuscular approach (Table II). These included four periprosthetic femoral fractures (at three weeks, six weeks, three months, and five months), each requiring stem revision and cerclage cable fixation, and two wound complications, each requiring superficial débridement and irrigation with primary wound closure. No other reoperations were performed, and there were no deep infections. There were five reoperations in the group treated with the less invasive direct lateral approach. One acetabular revision due to fracture at seven months was done in a patient who had initially presented with an acetabular fracture. In addition, there were two hematomas requiring irrigation and débridement, one superficial infection that resolved after irrigation and débridement, and one stitch granuloma requiring excision at two months. With the numbers studied, the rate of any component revision did not differ significantly between the two groups (four stems were revised in the group treated with the anterior supine intermuscular approach, and one cup was revised in the group treated with the less invasive direct lateral approach; p = 0.07).
Other notable complications in the group treated with the anterior supine intermuscular approach include two lateral femoral cutaneous nerve paresthesias, which both resolved (Table III). There were no dislocations, femoral nerve injuries, or cases of clinically evident deep-vein thrombosis. In the group treated with the less invasive direct lateral approach, one patient had paresthesias of the lateral aspect of the thigh and one had paresthesias related to the sciatic nerve, which eventually resolved. There were also three clinically evident deep-vein thromboses in this group.