Computer-assisted surgery accounts for pelvic orientation, thereby optimizing component positioning. Navigation has repeatedly been shown to decrease variability with regard to orientation of the acetabular component11-15 and to increase accuracy14,16-18, although there is a learning curve with its use18.
Computer-assisted navigation of the acetabular component first requires registration of anatomical landmarks so that the computer can determine where the pelvis lies in space. This is typically done by registering the anterior superior iliac spines and the pubic tubercle. By referencing these landmarks, the anterior pelvic plane is created, which is used for referencing cup position. For imageless navigation systems, registration is accomplished with optical trackers mounted to the pelvis. An optical pointer is then used to register the anterior superior iliac spines and the pubic tubercle either through small incisions or simply by palpation and then registration of the soft tissue directly over the anatomical landmark. For imageless systems, the registration portion of the process is done while the patient is supine in order to access the opposite anterior superior iliac spine. The patient may then be turned to the lateral decubitus position, if desired, which can be a problem when there is an optical tracker mounted to the pelvis.
Fluoroscopically guided navigation systems were sought for the purpose of improving the accuracy of the registration process and, ultimately, the final position of the cup. Fluoroscopic navigation allows for registration of the anterior pelvic plane in the lateral position. A series of fluoroscopic images in multiple planes are acquired to reference the position of the pelvis. Tannast et al. investigated the accuracy of fluoroscopically guided placement of the acetabular cup and found that the variability could be improved for cup abduction but not version19. Stiehl et al. found similar limitations with fluoroscopic referencing of pelvic landmarks, which may lead to variability in cup anteversion20,21. Other limitations include bulky equipment in the operating room and the necessity for radiolucent tables. An obese abdomen can impede the correct projection, which may lead to referencing errors19.
Kalteis et al. studied acetabular component position when the component was implanted with use of a freehand technique, with use of computed tomography-based navigation, or with use of imageless navigation22. They determined that freehand placement of the components resulted in only fourteen of thirty components being positioned within the safe zone as defined by Lewinnek et al. Twenty-five of thirty acetabular components in the computed tomography-based group and twenty-eight of thirty acetabular components in the imageless navigation group were positioned within this limit. Both computed tomography-based and imageless navigation techniques proved reliable22.
This single-surgeon series consists of a retrospective review of a computer database (CaptureWare; DePuy, Warsaw, Indiana) of all of the surgeon's patients. Between 2003 and 2006, the senior author (M.L.S.) performed 450 computer-assisted total hip arthroplasties. One hundred and twenty-five were performed with the aid of a computed tomography-based system, and 325 were performed with the aid of an imageless system. The goal for acetabular component placement was to reproduce the patient's anatomy on the basis of the acetabular inlet. Postoperative radiographs were taken at three months, and the measurements were recorded in the database for retrospective review.
Three hundred and thirty-five of the 450 computer-assisted total hip arthroplasties were performed with use of a minimally invasive approach. The results of these procedures compared favorably with the results of 192 total hip arthroplasties that were performed with use of conventional guides by the senior author during a time period between 2001 and 2004. In the minimally invasive, computer-assisted surgery group, the mean inclination was 48° (range, 32° to 59°) and the mean anteversion was 16° (range, 6° to 34°). There were three dislocations (<1%). The transfusion rate was 3%.
In the conventional group, the mean inclination was 46° (range, 30° to 60°) and the mean anteversion was 16° (range, 6° to 38°). There were three dislocations (2%), and the transfusion rate was 11%.
In a subset of the minimally invasive, computed tomography-based group, intraoperative versus postoperative radiographic measurements of acetabular component position were examined. Forty-two consecutive patients who underwent minimally invasive, computed tomography-based intraoperative navigation had a mean intraoperative acetabular component inclination of 44° ± 6.5°. The mean radiographic acetabular component inclination was 43° ± 5.8°.
Intraoperative measurements correlated well with postoperative radiographic measurements. Ninety-five percent of the intraoperative acetabular component measurements were within 10° of the postoperative radiographic measurement, 90% were within 8°, 75% were within 5°, and 48% were within 2°.
The mean intraoperative acetabular anteversion was 24° ± 12°. The range of anteversion measurements was consistent with the inclination measurements. However, postoperative radiographs were inconsistent with intraoperative measurements.
As previously discussed, the true plane of the acetabulum may be difficult to ascertain on the operating-room table. Pelvic tilt and rotation increased the variables for proper positioning of the acetabular component in relation to the anterior pelvic plane. In May of 2006, the senior author began to use the transverse acetabular ligament as a reference point to assist in determining the true version plane of the acetabulum. The transverse acetabular ligament is the part of the acetabular labrum that bridges the acetabular notch. When the transverse acetabular ligament is registered along with the superior aspect of the labrum, the result is the creation of a true acetabular inlet plane as a target for component placement (Fig. 1). Of eighty-seven hips navigated with this method, the average intraoperative abduction angle was 46° (range, 28° to 67°) and the average intraoperative anteversion angle was 18° (range, -2° to 44°) (Fig. 2). The average postoperative radiographic abduction angle was 39° (range, 26° to 56°) and anteversion was 20° (range, 14° to 36°). With use of the transverse acetabular ligament as a reference, 82% of the acetabular components were placed within the abduction safe zone and 71% were placed within the anteversion safe zone.
Postoperative limb-length inequality remains a major cause of patient dissatisfaction and litigation after hip replacement. Navigation has shown potential to help quantify intraoperative limb length and potentially reduce the risk for postoperative limb-length inequality (Fig. 3).
Component positioning is just as critical in hip resurfacing procedures. In one series, seventy-one consecutive hip resurfacing prostheses were placed with computer navigation. No significant difference was found between the intraoperative and postoperative cup inclination angles. There was a significant difference between the intraoperative and postoperative femoral neck-shaft angles (p < 0.001). The intraoperative neck-shaft angle was 142° ± 4° (range, 135° to 148°), while the radiographic neck-shaft angle was 146° ± 6° (range, 132° to 156°). Despite the learning curve for the procedure, computer-assisted navigation produced consistent values with regard to intraoperative cup inclination, postoperative radiographic alignment of the cup, and femoral neck-shaft angles27.
The use of navigation for total hip arthroplasty does add cost to the procedure as well as additional surgical time. There are obstacles to performing computer-assisted surgery. Additional equipment (for purchase or lease) is needed in the operating room, and additional navigation steps throughout the procedure add time to the case. Surgeons must also carefully consider which surgical approach to use, as computer-assisted imageless techniques require registration with the patient lying supine. Surgeons may need to reposition the patient after registration of the pelvis if the lateral decubitus position is desired.
However, recent literature has demonstrated that the percentage of acetabular components being placed in the target safe zone was higher in computer-navigated hips than in non-navigated controls, with less variation in implant position11-16. Even with a minimally invasive approach, the series presented here compares favorably with the conventional group. The results have also shown that intraoperative computer-assisted measurements of component inclination correlate well with the measurements made on postoperative radiographs. Intraoperative measurements were within 10° of postoperative radiographic measurements 95% of the time. This surpasses the results previously obtained with mechanical alignment guides9.
Studies by Lembeck et al.28 and Spencer et al.29 question the accuracy of computer navigation when the anterior pelvic plane is used as a reference to place the acetabular component. However, the use of computer-assisted surgery to identify the anterior pelvic plane has been associated with improved component position as compared with the component position achieved with use of mechanical guides12, or components placed freehand16. In addition, computer registration of the superior rim of the acetabulum along with the transverse acetabular ligament and subsequent adoption of that plane for acetabular orientation resulted in excellent radiographic alignment and a low postoperative dislocation rate in our patients. Registration of the transverse acetabular ligament and the superior acetabular rim may eliminate the need for registration of the anterior pelvic plane.
For hip resurfacing procedures, computer-assisted navigation is a dependable and accurate tool for the positioning of components to achieve the appropriate amount of acetabular inclination. Computer-assisted surgery reliably avoids varus stem placement and notching. Furthermore, computer navigation allows for consistent alignment independent of surgeon experience with hip resurfacing procedures27.
Computer-assisted surgery has contributed to reproducible and accurate positioning of hip arthroplasty implants. Computer navigation for minimally invasive approaches as well as hip resurfacing continues to evolve. To date, there are few studies that show functional improvement after total hip arthroplasty performed with computer-assisted surgery as compared with that after conventional total hip arthroplasty; however, with continued use of computer-assisted surgery, long-term studies may show a considerable beneficial effect as well as increased implant survivorship. Initially, hospitals may be hesitant to invest in a computer navigation system for hip arthroplasty. Taking into account the economic considerations of long-term implant survivorship, computer-assisted total hip arthroplasty may become the standard of care. 