Comparative Study of Acetabular Center Axis vs Anterior Pelvic Plane Registration Technique in Navigated Hip Arthroplasty

Abstract

There is significant variation in registering the anterior pelvic plane (APP) among experienced navigated hip arthroplasty surgeons, reflecting negatively on the accuracy of determining inclination and anteversion. Whether it is variation in pelvic anatomy or improper positioning, this inaccuracy emphasizes the need for alternative methods of registration, of which the acetabular center axis (ACA) is proposed. Data collected from ACA and APP registration were compared with postoperative computed tomography (CT) images of the pelvis in 34 cases. Findings showed ACA software to be comparable with CT in its accuracy in determining the inclination and version angles of the acetabulum and cup implant.

Currently, the anterior pelvic plane (APP) is used to identify the cup and acetabular orientation when navigating total hip arthroplasty (THA). First described by Cunningham¹ in 1922, the APP is based on the two anterior superior iliac spines (ASIS) and the two pubic tubercles. Recently, Jaramaz et al² have introduced the APP concept to computer-assisted cup placement in hip arthroplasty, and it has proved to be a useful tool.³ However, the reliability of APP registration as a reference system in a lateral decubitus position^4,5 is jeopardized because the contralateral ASIS is not readily accessible with either a pointer or the ultrasound methods.⁶ Furthermore, variation in thickness of subcutaneous tissue, the movement during the registration process, and the anatomic variations of acetabular version among healthy individuals resulted in major errors in cup orientation.⁷

To address these concerns, we postulate that the acetabular center axis (ACA) software is patient-specific, independent of variations in anatomy or pelvic position, and relies on readily accessible anatomic landmarks of the acetabulum rather than the anterior pelvic plane points. In this CT-based study, the reliability of ACA in determining acetabular anteversion and inclination angles is compared with that of the APP in computer navigation of THA.

Materials and Methods

This prospective study compares, through postoperative pelvis CT, the ACA registration with that of APP using anterolateral intermuscular mini-invasive computer assisted THA. Of the 36 prospective patients enrolled, 2 were excluded for lack of complete data. Patient age ranged from 34 to 83 years (mean, 63 years), and 31 patients were men. Twenty-six had primary osteoarthritis, and 8 had avascular necrosis. Mean body mass index was 29.2 kg/m ². Twenty-six percent of patients had dysplastic acetabulum, whereas 15% were protrusio.

Figure 1: APP plane.

The APP registration was done by palpation of both the ASIS and symphysis pubis at equal distance from the skin with the patient in the lateral decubitus position (Figure 1). Extra care was taken to ensure the stability of the pelvis during registration by stabilizing the pelvis with three posts (pegs) positioned anterior to and three pegs positioned posterior to the pelvis (Figure 2). The ACA registration was done by three points of palpation of the superior rim, three points of the most anterior rim, and three points of the most posterior rim of the acetabulum. The superior point must be chosen carefully so that the pelvic tilt can be kept constant. We determined the superior point of the acetabulum by palpating the iliac tubercle. A line drawn from the iliac tubercle to the center of the transverse ligament will cross the superior rim of the acetabulum at the desired point of registration. The computer will average the points (Figure 3) and map out the anatomy and orientation of the acetabulum, which will determine the acetabular center axis (ACA). This axis will guide the surgeon as to where to position the reamer for placement at the center of the acetabulum. The surgeon will have the option of choosing the desired version or inclination angle to accommodate individual patient anatomy.

Figure 2: Pelvis position secured.

The ACA and APP points were defined and registered in 34 consecutive patients who underwent OrthoPilot (B. Braun Aesculap, Tuttlingen, Germany) navigated press-fit Excia (B. Braun Aesculap) THA. To avoid common pitfalls of APP registration, care was taken to ensure that the registration of the APP plane was as accurate as possible by carefully securing the positioning of the patient and ensuring the equidistance between the registration point on the skin and the APP bony landmarks.

Preoperative plain pelvis radiographs with anteroposterior and cross-table lateral views were taken as routine. They were used to template the expected cup size in comparison with acetabular anatomy. Postoperative CT measurements of acetabular and cup inclination and version angles were observed independently using special software. Comparison of data was achieved using Fisher test and Student t test . The level of significance between CT and the variable should approach 1.0, which means the variable is as good as CT, when P < .05, the variable is significantly inferior to CT.

Results

Of the 36 consecutive patients, 34 were eligible for analysis of their hip data.

Mean anatomic (CT) acetabular version (ie, control) was 18.2º (SD ± 5.8), compared with 17.9º (SD ± 7.9) with ACA software. The mean anatomic (CT) acetabular inclination was 47.56º (SD ± 10.7). This reflects the validity and reliability of ACA software in identifying the version and inclination of the acetabulum.

Figure 3: A, ACA superior. B, ACA anterior. C, ACA posterior.

Cup implant version (CT) was 22.97° (SD ± 9.4), compared with 23.0° (SD ± 8.4) for the ACA software and 12.7° (SD ± 12.1) for the APP software. This reflects reliably and statistical superiority of ACA software in identifying the version of the cup implant ( P = .98), whereas the P value for the APP was significantly inferior to that for CT (P = .0002).

We then divided the patients into three groups according to anatomic variations of the acetabulum (normal, protrusio, and dysplastic). In the first group, the size of the cup closely matched the size of the acetabulum (normal acetabulum). The anatomic (CT) cup version was 21.7° (SD ± 10.3), compared with 21.7° (SD ± 8.8) for the ACA software (P = 1.0) and 11.37° (SD ± 10.5) for the APP software (P = .003). ACA software produced results identical to CT; however, the APP software was significantly inferior to CT.

In the second group, the size of the cup implant was smaller than the acetabulum (as in protrusio hips or in acetabulum with large osteophytes). The anatomic cup (CT) version was 22.0° (SD ± 9.9), compared with 22.8° (SD ± 9.5) for the ACA software and 14.3° (SD ± 15.9) for the APP software. Interestingly, both ACA and APP software were not statistically different in the protrusio group (P = .89 and .38, respectively).

In the third group, the size of the cup was larger than the acetabulum (as in dysplastic hips or the cup implant was larger by choice). The anatomic (CT) version was 26.3° (SD ± 7.1), compared with 26.5° (SD ± 6.4) for ACA software and 14.9º (SD ± 14.3) for APP software. Again, ACA software was as accurate as CT (P = .96), whereas APP software was less accurate (P = .04). Finally, when we compared the accuracy of detecting the version of the cup implant between ACA and APP software, there was a statistical difference between the two (P = .0001).

As for the inclination angle of the cup implant, mean anatomic (CT) cup inclination angle for all groups was 43.5° (SD ± 4.2), compared with 43.5° (SD ± 7.5) for the ACA software and 41.1° (SD ± 4.7) for the APP software. Both ACA software (P = 1.0) and APP software (P = .44) were accurate in detecting the inclination angle of the cup.

Similarly, we divided the patients into three groups for cup inclination comparisons. In the first group, the cup matched the acetabulum, the anatomic cup inclination angle was 42.7° (SD ± 3.6), ACA inclination was 43.1° (SD ± 4.7) (P = .73), and APP inclination was 40.4° (SD ± 4.7) (P = .097). Again, both ACA software and APP software were accurate. In the second group, the cup size was smaller than the acetabulum (representing protrusio), the CT-scan inclination angle was 42.6° (SD ± 4.0), ACA inclination was 46.8° (SD ± 6.6), and APP inclination was 42.2° (SD ± 4.0). Both software were accurate (P = .92). In the third group (representing dysplastic), the cup size was larger than the acetabulum, the anatomic cup inclination angle was 46.0° (SD ± 4.8), ACA inclination was 42.7 (SD ± 12.2) (P = .45), and APP inclination was 42.0° (SD ± 5.5) (P = .12). There was no statistical difference between ACA and APP software in detecting inclination angle in all types of acetabulae (normal, protrusio, or dysplastic) (P = .11).

In conclusion, ACA software was statistically superior to APP software in detecting the version of the cup. There was no statistical difference in the accuracy of the inclination angles between APP and ACA software. Both methods were within safety zone of Lewinnek.

Discussion

The anterior pelvic plane has been the corner stone of image-based hip navigation technologies. Cup orientation is usually defined by referencing the anterior pelvic plane (APP).^4,7-9 However, the APP does not consistently represent the functional pelvic position, and a small error in correctly identifying this plane results in a significant error in cup placement. Consequently, cup position parameters are not patient specific.¹⁰ Some centers recommended ultrasonography to identify APP with higher accuracy.⁶ However, this has been done in the supine position and, unfortunately, the accuracy was diminished in the lateral decubitus position. Other centers recommend acquiring the APP coordinate system landmarks in the supine position before turning the patient to the side, but this can be impractical and increases operative time with possible compromise to sterility. We conducted this study with the patient in the lateral decubitus position despite the limitation of the APP registration process in accessing the opposite ASIS. We attempted to overcome this limitation by measuring the distance between the palpation points and the bony landmarks to be registered with a ruler. We made every effort to make such distance equal and to be secure by padding or adjusting the pelvic position. Also, the movement of the pelvis was minimized by placing three pegs anteriorly and three pegs posteriorly to secure the pelvis.

The reliability of the OrthoPilot navigation system has been tested,⁶ and special software that independently reads both the APP and ACA registration points was developed and the data compared with postoperative CT of the pelvis as a control. Similar to the APP software, ACA software can accurately identify the cup orientation intraoperatively. However, ACA software supersedes APP software, which is easier to register and is independent of the pelvic position (movement during registration) and normal variations in anatomy. CT of the pelvis revealed significant variation of normal acetabulum version anatomy ranging from 5° to 30° of anteversion, where as the anatomic CT inclination angle range was less variable (47° ± 4°). This puts the safety zone of Lewinnek¹¹ in question. For example, a cup version of 29° is outside the safety zone of Lewinnek, but it is the normal acetabular version anatomy of a specific patient. For these variations, one study¹² demonstrated that there is perhaps no ideal position for the cup (45° inclination and 20° anteversion) that can be used for all patients. Because of the wide range of inclination and anteversion figures, half of cases in the study were outside the safety zone recommended by Lewinnek. This is another reason we believe ACA software to be superior to APP. ACA is more patient specific. The usefulness of the ACA concept to determine the orientation of the normal acetabulum has been well documented by Murray.¹³

Is there a limitation to the usefulness of the ACA software if there is a pathological variation of the acetabular anatomy (protrusio, dysplastic, or large osteophytes)? Will placing the cup implant in the ACA of the acetabulum re-create the original deformity and thus misplace the cup implant? That problem was resolved before the study began by obtaining standard pelvis A/P and cross-table lateral views preoperatively. If the acetabulum is determined to be dysplastic or the desired cup is larger than the acetabulum, templating will give an approximate inclination angle of the acetabulum and the cup. In protrusio, the acetabular inclination angle (<40°) is expected to be smaller than cup inclination angle, whereas in dysplastic cases (more than 50°), the acetabular inclination angle is expected to be larger than cup inclination angle. The difference in degrees is calculated and used intraoperatively to adjust the computer-navigated ACA measurement to the desired angle.

The computer will read the final cup position in the “normal” acetabulum anatomy as zero degrees. This means the cup should match exactly the individual acetabular anatomy. If the acetabulum is protrusio, the computer readings of inclination should be in the +ve (positive) range, which means that the cup will be smaller than the acetabulum. If the acetabulum is dysplastic, the computer reading of the cup inclination should be in the –ve (negative) range, which means that the cup is larger than the acetabulum and that some of the rim of the cup is outside the acetabulum. This will enable the surgeon to make the proper adjustment in grossly abnormal acetabulum to achieve the ideal cup position specific to each patient. However, no such adjustments are needed to determine the cup version.

Finally, this study shows that there are two techniques (APP and ACA) for referencing in the lateral position. They are statistically different in terms of accuracy in determining the cup version compared with CT. As for the inclination angle, both APP and ACA were accurate. However, our new reference axis (ACA) has the advantage of being patient specific and independent of variations in anatomy or pelvic position. The system relies on readily accessible anatomical landmarks of the acetabulum, making it significantly attractive for surgeons who use CT-free planning and navigation.

References

Cunningham DJ. Pelvis. In: Cunningham DJ, ed. Cunningham’s Textbook of Anatomy. London: Hodder & Stoughton; 1922:255-260.
Jaramaz B, DiGioia AM, Blackwell M, Nikou C. Computer assisted measurement of cup placement in total hip replacement. Clin Orthop Rel Res. 1998; (354):70-81.
Zheng G, Marx A, Langlotz U, Widmer KH, Buttaro M, Nolte LP. A hybrid CT-free navigation system for total hip arthroplasty. Comput Aided Surg. 2002; 7(3):129-145.
Archbold HA, Mockford B, Molloy D, McConway J, Ogonda L, Beverland D. The transverse acetabular ligament: an aid to orientation of the acetabular component during primary total hip replacement. J Bone Joint Surg Br. 2006; 88(7):883-886.
Wolf A, Digioia AM, Mor AB, Zheng G, Jaramaz B. Cup alignment error model for total hip arthroplasty. Clin Orthop Relat Res. 2005; (437):132-137.
Kieffer H, Othman A. Ultrasound vs pointer palpation based method in THA navigation: a comparative study. Orthopedics. 2007; 30 (suppl 10):S153-S156.
Spencer JM, Day RE, Sloan KE, Beaver RJ. Computer navigation of the acetabular component: a cadaver reliability study. J Bone Joint Surg Br. 2006; 88(7):972-975.
Ohashi H, Matsuura M, Okamoto Y, Ebara T, Kakeda K, Takahashi S. Status of navigated total hip arthroplasty in dysplastic osteoarthritis. Orthopedics. 2007; 30 (suppl 10):S117-120.
Tannast M, Langlotz U, Siebenrock KA, Wiese M, Bernsmann K, Langlotz F. Anatomic referencing of cup orientation in total hip arthroplasty. Clin Orthop Relat Res. 2005; (436):144-150.
Pinoit Y, May O, Girard J, Laffargue P, Ala Eddine T, Migaud H. Low accuracy of anterior pelvic plane to guide the position of the cup with imageless computer assistance: variation of position in 106 patients. Rev Chir Orthop Reparatrice Appar Mot. 2007; 93(5):455-460.
Lewinnek GE, Lewis JL, Tarr R, et al. Incidence of dislocation following hiparthroplasty for patients in rehabilitation setting. J Bone Joint Surg Am. 1978; 60(2):217-220.
Laffargue P, Pinoit Y, Tabutin J, Giraud F, Puget J, Migaud H. Computer-assisted positioning of the acetabular cup for total hip arthroplasty based on joint kinematics without prior imaging: preliminary results with computed tomographic assessment. Rev Chir Orthop Reparatrice Appar Mot. 2006; 92(4):316-325.
Murray DW. The definition and measurement of acetabular orientation. J Bone Joint Surg Br. 1993; 75(2):228-232.

Authors

Drs Hakki and Oliveira are from Bay Pines Health Care System in Bay Pines, Florida.

Dr. Hakki is a consultant for B. Braun Aesculap. Mr. Dordelly and Dr. Oliveira have no relevant financial relationships to disclose.

Correspondence should be addressed to: Sam Hakki, MD, PO Box 22429, St. Petersburg, FL. 33742.