Anatomy and Biomechanics of the Hip Relevant to Arthroplasty

Even though an enormous volume of literature concerns the anatomy and biomechanics of the hip, little of it is specifically organized from the perspective of total hip arthroplasty. Also, since an analysis of the finer elements of these factors has formed a significant part of the development of the porous coated anatomic (PCA) hip system, those aspects that we believe to be particularly important are reviewed in some detail.

Although a three-dimensional perspective is of critical importance in developing any kind of total hip system, we must recognize that much of the information that we gather in every day practice is two dimensional, namely, x-rays of the hip. The axial view provided by computed axial tomography (CAT) scanning is beginning to become a part of the common knowledge, but the information that it imparts has not yet had much impact on prosthetic design. The detailed cross-sectional anatomy presented in this chapter provides the informational background for the design of the PCA hip. It is our hope that this section will be particularly well studied.

The section on anatomy presents only the soft tissues that are encountered at hip arthroplasty from the perspective of the surgeon. However, since the direct lateral approach, which we are using, may be unfamiliar to many, these anatomical details will also be helpful for a better understanding of the surgical exposure.


The Proximal Femur

The form of the femur is relatively complex, with bows and twists that distort its basically tubular structure. The anterior bow of the midportion of the femur is well recognized and has even been built into some current prostheses. This is commonly envisioned as an anterior bow because of the position that the separate femur assumes when it is placed on a horizontal surface, resting on the posterior margin of the trochanter and the posterior aspects of the condyle (Figure 1.1A). However, in vivo the orientation is somewhat different. In the erect position, the central portion of the femur is more in the coronal plane of the body, with the distal portion inclined posteriorly to the knee and the proximal portion inclined anteriorly to the acetabulum (Figure 1.1B).

The posterior bow of the proximal femur is just as constant as the midportion anterior bow, but it seems to have been unrecognized or considered of no importance by most designers of femoral stems. The central portion of the proximal posterior bow is opposite the level of the lesser trochanter. This bow is constant. It is also noteworthy that the radius of curvature does not seem to change dramatically with the size of the femur. The length of the curve increases with increasing femur size from the base of the neck until the curve reverses distal to the lesser trochanter, but the radius of the curve is relatively constant.



Figure 1.1 A, Photo of skeletal femur resting on table top; B, same femur positioned in the neutral plane, as it is in the body.


The Neck-Shaft Angle

The head of the femur considerably overhangs the femoral shaft. This occurs because the neck makes an oblique angle with the shaft of an average of 135° (3). Although there is considerable variability in both the neck-shaft angle and neck length, in general the center of the femoral head is extended medially and proximally by the femoral neck so that the center of the femoral head is at the level of the tip of the trochanter. The effect of the overhanging head and neck is to lateralize the abductors, which attach to the greater trochanter, from the center of rotation (center of the femoral head). This increases the torque generated by the abductors and reduces the overall force necessary to balance the pelvis during single leg stance. Reducing this level arm (coxa valga) increases total load across the hip, and coxa vara reduces it to the extent it increases the lever arm. (Coxa vara with a short neck would have a negative affect.)

Figure 1.2 Photograph of a femur viewed from the end, showing anteversion of the head and neck.

Femoral Anteversion

The coronal plane of the femur is generally referenced to the posterior distal femoral condyles. When oriented in this plane, it can be seen that the proximal femur, including the femoral head and neck, are rotated anteriorly. This is commonly referred to as femoral head-neck anteversion. However, it is really a combination of a torsional change in the intertrochanteric part of the femur and a further anteversion of the femoral neck based upon this torsion. The sum of this change is that the adult femoral head and neck are in a plane 10-15° anteriorly oriented to the coronal plane (3, 5; Figure 1.2).

This can be very different in pathologic cases, particularly hip dysplasia and congenital hip displacement where torsional changes can be as much as 80°. Such abnormal relationships are much more frequently a true torsion rather than an abnormal relationship between the femoral head, neck, and greater and lesser trochanters. This is an important distinction when considering what changes may be indicated in handling such patients who present for total hip replacement. For prostheses intended to achieve maximal purchase in the proximal femur, it may be necessary to first change the relationship between the proximal and the distal femur.

It is important to view the hip three dimensionally in order to establish criteria for forming prosthesis geometry that is anatomical. Both CAT scans of the proximal femur in normal patients and transection of properly oriented, referenced, and marked cadaver femurs can assist in this understanding. We feel that the intertrochanteric region of the hip is important for long-term fixation and stability, while the diaphyseal region distal to the lesser trochanter is principally important for initial fixation and stability of the components. Therefore our analysis will concentrate in these regions.

Figure 1.3 A-P x-ray of femur showing the well developed compression and tension trabecular patterns of the proximal femurs.
Figure 1.4 Coronal cut through the proximal femur showing the distribution of cancellous bone and particularly the striking diminution distal to the lesser trochanter.

Distribution of Cancellous Bone in the Proximal Femur

A critical look at a good quality anteroposterior (A-P) x-ray of the femur gives a good idea of the distribution of cancellous bone in the femur (Figure 1.3). Even though the orientation of the trabecular pattern may be significantly disturbed in the diseased hip, the overall distribution of cancellous bone is still the same. It appears to be a characteristic of the articulating ends of long bones that the broad ends, covered with articular cartilage, are supported principally by cancellous bone and a very rudimentary cortex in the form of a subchondral plate. The forces applied to the articular surfaces are carried by the cancellous bone out to the cortex. It does not appear to be a coincidence that where the cortex reaches its full thickness, the cancellous bone essentially stops. The distribution of cancellous bone that is suggested in the x-ray is vividly illustrated in the coronal cut through a dessicated femur (Figure 1.4).

Figure 1.5. Direct lateral x-ray of the femur. Arrows mark the structures that limit the windows of access for a straight-stem prosthesis.

Cross-Sectional Analysis

As a part of the development of the shape and sizes for the PCA hip stem, we x-rayed 86 North American cadaver femurs in the A-P and true lateral position to determine the form of the proximal femur and distribution of cancellous bone. A detailed examination of one of those normal bones is representative for the shape of the normal human femur. Particularly on the lateral x-ray, the posterior bow of the proximal femur can be seen with its apex opposite the lesser trochanter (Figure 1.5). The three aspects of the anatomy of the femur that limit the access of stems that are straight in the lateral plane are the posterior margin of the femoral neck, the anterior margin of the cortex opposite the lesser trochanter, which represents the apex of the posterior bow of the femur, and the posterior cortex of the shaft where the bow of the femur is reversing into an anterior bow.

Figure 1.6. A-P x-ray of the proximal femur with overlay indicatinng the levels of transection.
Figure 1.7. Slice 3 through the tip of the trochanter and center of the femoral head. The constant horizontal line represents the coronal plane in this and subsequent x-rays. Total anteversion is 13°.
Figure 1.8. Slice 5. Arrow marks the posterior margin of the femoral neck. Its anterior level blocks the sinking of a straight stem prosthesis posteriorly to get past the bow in the femur at the level of the lesser trochanter.
Figure 1.9. Slice 7. Note the dense cancellous bone in proximity to the relatively thin cortex.

The femur was sliced into 30 cross-sectional views that then were subjected to contact radiography to show the form of the cortical outline and the distribution of the cancellous bone. The levels of the cuts and their numbers are shown on an A-P x-ray (Figure 1.6). From slice 3 at the tip of the greater trochanter, it can be seen that this corresponds to the center of the femoral head (Figure 1.7). At slice 5 the superior margin of the femoral neck is transected, and it can be seen how far the posterior margin of the neck is anteriorly oriented (Figure 1.8). Slice 7 in the intertrochanteric region shows the distribution of the cancellous bone and particularly the density of the trabecular pattern near the regions of the cortex, which itself is relatively thin (Figure 1.9). In slice 10 through the calcar, the trabecular pattern is somewhat looser, but again clearly dense within 2 or 3 mm of the cortex all around the cortical margin, with the exception of the posterior intertrochanteric ridge (Figure 1.10). Slice 13 represents the area approximately 1 cm above the lesser trochanter, and again the density of the cancellous bone can be seen to follow the cortical outline (Figure 1.11). Slice 15 is through the lesser trochanter, and between slice 15 and slice 17, a dramatic change in the density of the cancellous bone can be seen (Figures 1.12 and 1.13). When comparing this back to the A-P x-ray (see Figure 1.6), the compression-oriented trabeculae, which are located in the medial aspect of the head and neck and insert into the calcar, can be seen to end at the level of the lesser trochanter, and the tension trabeculae, which arc from the lateral trochanter into the head and intersect the compression trabeculae at nearly right angles, can also be seen to end at the same level. Therefore, the distribution of cancellous bone is limited to an area proximal to the inferior margin of the lesser trochanter and the lateral flare at the base of the greater trochanter. Below this area there are only a few coarse trabecular filaments in direct apposition to the cortical margins.

When comparing slice 13 with slice 17, it can be seen that there is a dramatic change in the density of the cortical outline itself; so it appears that as the trabeculae carry the stress progressively out to the cortex, the cortex gradually takes over, and the cancellous bone is no longer needed as a vehicle for stress transfer. This carefully oriented transectional analysis of the proximal femur also shows the rotational orientation of various levels of the femur. The constant transverse line represents the coronal plane. At the level of slice 20, it can be seen that the femoral shaft is basically a circular tube with no particular rotational orientation (Figure 1.14). However, at slice 15, which is through the level of the lesser trochanter, the anterior cortex already can be seen to be anteverted a few degrees. This degree of anteversion has increased slightly at slice 13. The degree of twist of the proximal femur is a little more difficult to evaluate proximal to slice 12 since the intertrochanteric area is irregularly shaped. However, in looking at transection at level 3, which is at the tip of the trochanter, approximately 13° of femoral anteversion can be seen in this specimen. The anteversion has already begun at the level of the lesser trochanter, continues throughout the intertrochanteric region, and is completed in the anteversion of the femoral neck.

By superimposing the cross sections from the proximal base of the neck, the level of the lesser trochanter, and the level of the tip of a typical prosthesis, the limit to the window of common opening for a straight stem prosthesis can be seen (Figure 1.15). The straight stem would bind proximally at the posterior margin of the neck, in the mid-portion at the anterior cortex, and distally at the posterior cortex. A larger stem prosthesis would have the tendency to blow out the posterior neck as the stem follows the anterior bow of the midfemur or to punch through the posterior cortex 5-6 in. down the shaft.

Although some of the pathology of the adult hip is a result of abnormal anatomy, the vast majority of patients presenting for total hip replacement have degenerative changes superimposed on a relatively normal anatomy. Therefore it should be possible to fit the majority of the patients who need total hip replacement with prostheses based on a careful analysis of the form and cancellous bone distribution of the normal human femur.

Figure 1.10. Slice 10. The density of the cancellous bone parallels the cortical outline, with the exception of the posterior intertrochanteric ridge.
Figure 1.11. Figure 13, 1 cm above the lesser trochanter. The anterior cortex shows some anteversion in relation to the coronal plane.
Figure 1.12. Slice 15 through the lesser trochanter, showing hte end of the dense cancellous bone, thickening of the cortex, and the beginning of anteversion.
Figure 1.13. Slice 17. The cortex is now fully developed and the cancellous bone is rudimentary.
Figure 1.14. Slice 20. the diaphysis is essentially a cortical tube with little rotational orientation and only a small amount of cancellous bone.


The acetabulum is formed by the confluence of the ilium, pubis, and ischium at the triradiate cartilage. With the fusion of this cartilage at the completion of growth, the form and orientation of the acetabulum are fixed. The normal bony acetabulum is slightly less than a hemisphere, but its functional dimensions are extended by the tough fibrocartilaginous labrum. There is some controversy in the literature concerning the orientation of the acetabulum. Getz (2), Steindler (9), and von Lanz (11) all give figures for acetabular anteversion of around 4O° (38- 42°). They made their measurements with the pelvic brim approximately horizontal, where it can be seen that the acetabula are obviously facing forwards (Figure 1.16). However, in the erect position, the anterior-superior iliac spines and pubic symphysis are in the same plane and the acetabulae are not as obviously anteverted (Figure 1.17). McKibbin measured 30 each adult male and female pelvises, oriented this position, and recorded an average anteversion of 14° (5-19°) for men and 19° (10-24°) for women (6). When the pelvis is flexed, as it is in sitting, the forward facing of the acetabulum is accentuated. In the erect position, the anterior-superior iliac spines and the symphysis pubis lie in the coronal plane. In this position, the acetabulum opening is directed approximately 45° laterally and 15° forward. Flexion of the pelvis on the lumbar spine increases the apparent anteversion without greatly changing the apparent lateral opening, whereas lumbar lordosis does the opposite.

Figure 1.15. Superimposition of slices 6, 15 and 26 give an end-on view of the proximal femur. The common A-P window for a straight stem prosthesis is narrowed by the posterior margin of the neck (slice 6), the anterior cortex of slice 15, the posterior cortex of slice 26. The slices have been oriented against a common coronal plane reference.
Figure 1.16. With the pelvis positioned so that the brim is nearly horizontal, the acetabula obviously face forward.
Figure 1.17. In the neutral position (anterior-superior iliac spines and pubic symphysis all in the coronal plane) the acetabular anteversion is less marked.
Figure 1.18. Standard A-P x-ray of the hip suggests that the body of the ilium is very broad. Acetabular rim is outlined with malleable wire.

Because of the oblique orientation of the acetabulum to the coronal plane, an A-P view of the pelvis gives a view of the acetabulum that does not give a true picture (Figure 1.18). Judet internal and external rotation views coupled with the A-P and the cross-table lateral x-rays are necessary to give a better representation of this complex structure (Figures 1.19-21). As with the femur, transverse sections and CAT scan sections of the acetabulum are helpful in constructing the accurate three-dimensional image that is so important for correct component placement at total hip arthroplasty (Figure 1.22).

Figure 1.19. Internal Judet view shows the body of the ilium in its true thickness and an end-on view of the acetabulum.
Figure 1.20. External Judet view shows the width of the body of the ilium and the anterior and posterior margins of the acetabulum overlapping.
Figure 1.21. Cross-table lateral x-ray shows the forward facing acetabulum.



Figure 1.22. A, CAT scan of the acetabulum through the midsection shows apparent anteversion. B, More proximal cut. Compare thickness of body of the ilium to "apparent" thickness in Figure 1.18 and true thickness in Figure 1.20.
Figure 1.23. Schematic showing the direction and magnitude of the load on the femoral head in symmetrical two-leg stance. [Redrawn from Pauwels F. Biomechanics of the Locomotor Apparatus. springer Verlag, New York, 1980 (7).]

Biomechanics of the Hip

This section is not intended to be a comprehensive analysis of the forces acting on the proximal femur and the acetabulum. The reader is referred to the exhaustively complete analysis of this subject by Frederick Pauwels in Biomechanics of the Locomotor Apparatus (7). It is, however, important to the success of total hip arthroplasty that one understands the factors influencing both the direction and magnitude of forces acting upon the femoral head. The forces exerted on the hip have their biological expression in the form of the femur and acetabulum, particularly in the location and orientation of the trabecular pattern. The forces exerted on the prosthetic femoral head in a properly performed total hip replacement will be very similar in both direction and magnitude.

Of all the species in the animal kingdom, only birds and man habitually use a bipedal gait. Even the larger primates use a quadripedal ambulation mode for most of their activity. When the weight of the body is being borne on both legs, the center of gravity is centered between the two hips and its force is exerted equally on both hips (Figure 1.23). Under these loading conditions, the weight of the body minus the weight of both legs is supported equally on the femoral heads, and the resultant vectors are vertical.

When the hips are viewed in the sagittal plane and if the center of gravity is directly over the centers of the femoral heads, no muscular forces are required to maintain the equilibrium position, although minimal muscle forces will be necessary to maintain balance. If the upper body is leaned slightly posteriorly so that the center of gravity comes to lie posterior to the centers of the femoral heads, the anterior hip capsule will become tight, so that stability will be produced by the Y ligament of Bigelow. Therefore, in symmetrical standing on both lower extremities, the compressive forces acting on each femoral head represent approximately one-third of body weight (7).

In a single leg stance, the effective center of gravity moves distally and away from the supporting leg since the nonsupporting leg is now calculated as part of the body mass acting upon the weight-bearing hip. Since the pillar of support is eccentric to the line of action of the center of gravity, body weight will exert a turning motion around the center of the femoral head. This turning motion must be offset by the combined abductor forces inserted into the lateral femur. In the erect position, this muscle group includes the upper fibers of the gluteus maximus, the tensor fascia lata, the gluteus medius and minimus, and the pyriformis and obturator internus. The combined resultant vector of the abductor group can be represented by the line of action M in Figure 1.24. Since the effective lever arm of this resultant force (BO) is considerably shorter than the effective lever arm of body weight acting through the center of gravity (OC), the combined force of the abductors must be a multiple of body weight. The vectors of force K and force M produces a resultant compressive load on the femoral head that is oriented approximately 16° obliquely, laterally, and distally. The orientation of this resultant vector is exactly parallel to the orientation of the trabecular pattern in the femoral head and neck (Figure 1.25).

Figure 1.24. Forces of the hip in single leg stance. G, Center of gravity; M, muscle forces; K, effect of partial body weight; R, resultant vector. [Redrawn from Pauwels F. Biomechanics of the Locomotor Apparatus, Springer Verlag, New York, 1980 (7).]
Figure 1.25. A-P x-ray of a normal hip showing the compression trabeculae oriented parallel to the resultant compressive load on the femoral head.

The effect of this combined loading of body weight and the abductor muscle response required for equilibrium results in the loading of the femoral head to approximately 4 times body weight during the single leg stance phase of gait. This means that in normal walking the hip is subjected to wide swings of compressive loading from one-third of body weight in the double support phase of gait to 4 times body weight during the single leg support phase. The factors influencing both the magnitude and the direction of the compressive forces acting on the femoral head are 1) the position of the center of gravity; 2) the abductor lever arm, which is a function of the neck-shaft angle; and 3) the magnitude of body weight. Shortening of the abductor lever arm through coxa valga or excessive femoral anteversion will result in increased abductor demand and therefore increased joint loading. If the lever arm is so shortened that the muscles are overpowered, then either a gluteus minus lurch (the center of gravity is brought laterally over the supporting hip) or a pelvic tilt (Treridelenburg gait) will occur.

Figure 1.26.  Forces on the hip with sideways limping. Note the reduction of vector M and R even though K is unchanged. R is also more vertically oriented. [Redrawn from Pauwels F. Biomechanics of the Locomotor Apparatus, Springer Verlag, New York, 1980 (7).]

Since the loading of the hip in the single leg stance phase of gait is a multiple of body weight, increases in body weight will have a particularly deleterious effect on the total compressive forces applied to the joint. The effective loading of the joint can be significantly reduced by bringing the center of gravity closer to the center of the femoral head (Figure 1.26). Sideways limping, however, requires acceleration of the body mass laterally, its deceleration during the stance phase of gait, and then its acceleration back to the midline or even to the other side as the single leg stance phase changes to the opposite extremity. This requires considerable energy consumption and is a much less efficient means of ambulation than the normal situation in which the hip is subjected to these considerable forces. Another effect of sideways limping is that the resultant vector becomes more vertical because the center of gravity is acting in a more vertical direction, and therefore the bending moment the femoral neck is increased.

Figure 1.27. Forces on the hip with the use of a cane. Force levels and vectors correspond to the magnitude of pressure referred to in Table 1.1. [Redrawn from Pauwels F. Biomechanics of the Locomotor Apparatus. Springer Verlag, New York, 1980 (7).]

Another mechanism for reducing the resultant load on the femoral head is the use of a walking stick in the opposite hand. Since some of its force is transferred to the walking stick through the hand, the effective load of body weight is thus reduced in two ways: 1) the effective load of body weight is reduced; 2) since the turning moment around the femoral head is reduced, the abductor demand is also reduced (Figure 1.27).

TABLE 1.1. Influence of a Walking Stick on Forces across the Hip
Pressure of stock (kg)
Static load across the hip(kg)
Angle in inclination from the vertical of the compression force on the femoral head

Adapted from Pauwels F. Biomechanics of the Locomotor Apparatus. Springer Verlag, New York, pp 1-228, 1980 (7).

Pauwels (7) has calculated both the total compressive load on the femoral head and the angle of inclination of the vertical compressive loads for different forces applied to the walking stick (Table 1.1). It can be that only 9 kg of force applied to a cane in the opposite hand reduces the load on the femoral head by nearly 40%. The same effect could also be achieved by a 40% reduction in body weight. Also the angle of inclination with this degree of unloading is not significantly different from normal, so that using a stick to unload the femoral head produces lower bending forces around the femoral neck than sideways limping. Therefore, in the rehabilitation of patients after total hip arthroplasty, the use of a stick to prevent sideways limping is always preferable.

Figure 1.28. A-P and lateral photographs of three Charnley femoral stems: unused stem (a), failed prosthesis (b), fatiguing prosthesis (c); b and c are deformed in varus and retroversion.

The form of the femur and the orientation of the trabecular pattern in the proximal femoral metaphysis and epiphysis would support the conclusion that the principal loading of the femoral head is in the coronal plane. However, there is another manner of loading that also has clinical relevance to total hip arthroplasty and may also play a significant role in loosening. When an individual rises from the seated position or climbs stairs, the forces of body weight are applied to the anterior surface of the femoral head. The femur itself is prevented from rotating in response to this applied load by the stabilization of the posterior femoral condyles against the tibial plateaus. In addition the psoas tendon inserting into the lesser trochanter prevents this applied load from rotating the femur internally. This anteriorly applied force therefore produces a twisting strain on the proximal femur. That this must be so is demonstrated in two Charnley total hip femoral stems that were recovered after failure through loosening. In both instances the distal portion of the prosthesis remained fixed in the diaphysis while the proximal cement mantle loosened. Although both specimens had deformed into varus, they both also had deformed more in retroversion (Figure 1.28). The more deformed of the two specimens was from a 40-yr-old postal worker who had a total hip replacement for avascular necrosis and returned to work as a postman, which required frequent squatting and lifting of packages.

This aspect of loading of the proximal femur takes on particular importance for femoral stem design since anteriorly applied loads will produce a twisting strain on the stem within the medullary canal. Vertical loading of the femoral component will produce compressive load on the medial side of the femoral stem and tension loads on the lateral side of the stem, whereas anterior loading will produce shear stresses at the prosthesis-bone-cement interfaces. Since smooth stems are capable of transmitting load only in compression, this latter mode of loading is an argument for fixation that has the capability of transmitting all three mechanisms of stress: compressive, tensile, and shear. It also implies that it is inadequate to analyze the validity of femoral stem design by only simulating vertical load and that the resistance to twisting moments within the femoral canal also requires analysis.

Forces Acting on the Acetabulum

Many more detailed analyses of the biomechanics of the hip have been directed toward the stresses within the femoral stem than within the acetabulum. However, in the long-term follow-up of Charnley, acetabular loosening has been an important problem (1). The intact acetabulum is a horseshoe form that wraps around the superior, anterior, and posterior aspects of the slightly eccentric femoral head. In the lightly loaded state, the dome of the acetabulum is relatively unloaded, and the stress is transferred from the femoral head to the acetabulum through the anterior and posterior extensions of the horseshoe. As the load is progressively applied, since the acetabulum is not in continuity inferiorly, the anterior and posterior sides of the horseshoe are free to expand so that a more congruous seating of the femoral head is allowed. As Radin has pointed out, this phenomenon of deformation under load leads to increasing congruity with progressive loading (8). If the hip were fully congruent in the acetabulum, full loading would produce incongruence as the anterior and posterior extensions of the horseshoe would separate away from the femoral head on loading. This deformation of the acetabulum under load has relevance to total hip arthroplasty since loading of a deformable polyethylene cup could lead the polyethylene to separate from the acetabulum due to the deformability of both materials.

The analysis of the forces acting on the femur also apply to the acetabulum. The orientation of the resultant vector passing through the acetabulum should pass through the center of the body of the ilium (see Figure 1.2). If there is protrusio acetabuli, then this force will pass through the medial wall, which will ultimately fail with progression of the protrusio. If the vector is lateralized or the acetabulum dysplasic, subluxation and lateral acetabular hip erosion may occur.

Vasu, Carter, and Harris have analyzed the distribution of stresses in the acetabulum before and after total hip replacement, using finite element analysis (10). In the normal hip they found transmission of compressive stresses by the cancellous bone of the body of the ilium to the lateral acetabulum wall and lesser order tensile stresses to the medial wall. After conventional total hip replacement, the compressive stresses in the cancelbus bone immediately above the cup were increased, as well as tensile and compressive stresses in the medial wall. Stresses in the lateral wall were decreased. Adding metal backing to the cup redistributed the stresses throughout the whole acetabulum so that stress in the cancellous bone was reduced.

Figure 1.29. Lateral view of pelvis and principal muscle origins.

Figure 1.30. Anterior and posterior view of the proximal femur, showing principal muscle insertions.

Anatomy of the Soft Tissues About the Hip

An exhaustive review of all of the anatomical structures around the hip is beyond the scope of this monograph, but it is important to put into perspective some of the more important structures that are encountered through the standard surgical approaches for total hip arthroplasty. This is even more important because the direct lateral approach, which is our preferred approach, is not currently being extensively utilized in this country. The perspective of this approach is slightly different from other approaches with which the surgeon may be more familiar. Figures 1.29 and 1.30 show the pelvic origins and femoral insertion of the muscles described below.

Figure 1.31. The deltoid of the hip.

The first structure encountered at total hip arthroplasty after the incision of the skin is the fascia lata with its muscular inputs from the tensor fascia lata and the gluteus maximus (Figure 1.31). Kapandji has referred to this as the deltoid of the hip (4). The tensor fascia lata is a relatively small, narrow muscle that originates from the pelvis at the inferior margin of the crest just behind the anterior-superior iliac spine and runs posteriorly and inferiorly to insert into the fascia lata. It functions as a flexor and abductor of the hip. In combination with the gluteus maximus, the tensor serves to tense the iliotibial tract, which itself functions as a tension band in offsetting the bending forces that are applied to the femoral head. The tensor fascia lata is innervated by a branch of the superior gluteal nerve coming out from underneath the gluteus medius. This takes on some importance when carrying out the anterolateral approach to the hip, because carrying the dissection between the tensor fascia lata and the gluteus medius too proximally can result in denervation of the tensor fascia lata. The gluteus maximus is the largest and strongest muscle of the body. From its origin on the posterior third of the iliac crest and the dorsum of the sacrum and coccyx, it runs obliquely, inferiorly, and anteriorly to insert into the fascia lata and also into the posterolateral margin of the femur just below the level opposite the lesser trochanter. The superior fibers of the gluteus maximus function as abductors and contribute to the tension in the iliotibial tract. The main body of the gluteus maximus, however, functions as a hip extensor. The conjoined tendon of the insertion of the gluteus maximus and the fascia lata into the femur at the linea aspera is an important anatomical landmark. Proximal to the conjoined tendon, the posterior aspect of the proximal femur is freely accessible. Distal to the conjoined tendon, from a lateral approach, the posterior compartment of the thigh is separated by insertion of the attachment of dense fascial tissue into the linea aspera. The innervation of the gluteus maximus is from the inferior gluteal nerve, which leaves the pelvis through the greater sciatic notch below the pyriformis. This would not normally be at risk at total hip arthroplasty.

Figure 1.32. Orientation of the principal hip abductors.

The next structures encountered at total hip arthroplasty, once the fascia lata has been split and retracted anteriorly and posteriorly, are the abductors (Figure 1.32). The most important of these is the gluteus medius, which originates from the wing of the ilium just below the crest. The origin for the gluteus medius extends across the whole breadth of the wing of the ilium, and the broad fan-shaped muscle narrows to a distal insertion on the lateral and anterior surfaces on the greater trochanter. The posterior margin of the gluteus medius is well defined by a thick tendon, which inserts into the tip of the trochanter just anterior to the pyriformis tendon insertion. This tendinous condensation of the gluteus medius will take on some importance in the description of the direct lateral approach to the hip. It is not always apparent by visual inspection since it is completely surrounded by muscle fibers, much as the psoas tendon is surrounded by the iliacus fibers at the level of the pelvic brim. However, it is readily palpable and is approximately the thickness of the thumb. The anterior fibers of the gluteus medius are less well defined since they are overlapped by the tensor fascia lata.

Figure 1.33. The pyriformis provides a critical landmark for exiting nerves and vessels.

The gluteus medius is innervated by branches from the superior gluteal nerve, which exits the pelvis through the greater sciatic notch above the pyriformis in the company of the superior gluteal vessels (Figure 1.33). The superior gluteal nerve and vessels pass deep to the gluteus medius muscle mass distal to the crest of the wing of the ilium and superficial to the gluteus minimus muscle. The terminal branch of the superior gluteal nerve exits underneath the anterior margin of the gluteus medius muscle to enter the tensor fascia lata. Although this nerve would not be in jeopardy with the common approaches to the hip, one must be aware of its location because it forms the upper limit to the proximal extent of the incision in the gluteus medius muscle when using the direct lateral approach. It is also not possible to simply proximally extend the incision in the muscle to the level of the pelvis if one desires to expose the wing of the ilium as a source of bone graft. The next strongest abductor is the gluteus minimus, which originates from the wing of the ilium just beneath the gluteus medius. It, too, extends the full width of the wing of the ilium, in this case just anterior to the greater sciatic notch to the level of the bridge between the anterior-superior and anterior-inferior iliac spines. From this broad origin, it narrows sharply to insert onto the anterior-superior greater trochanter, deep and anterior to the insertion of the gluteus medius tendon. There is a small amount of fatty bursal tissue between these two tendinous insertions. When the hip is exposed by using the direct lateral approach, the posterior-most fibers of the gluteus medius may resist the anterior retraction, necessitating transection of that posteriormost portion of the tendon. The pyriformis muscle is also an abductor as well as an external rotator. It takes on particular importance at total hip surgery because it delineates the superior from the posterior portions of the exposure. When using the direct lateral approach, the pyriformis defines the posterior area for the short external rotators, which are not violated. When the posterior approach is used, the pyriformis delineates the superior extent of the muscle release and incision (Figure 1.33). The flat muscle belly of the pyriformis lies almost parallel to the posterior margin of the gluteus medius. It arises from the lateral margin of the anterior surface of the sacrum and the margin of the greater sciatic foramen, passing out of the pelvis through the greater sciatic foramen to insert into the tip of the greater trochanter. It is frequently blended at its insertion with the common tendon of the obturator internus and gemelli. The sciatic nerve passes deep to the pyriformis. The nerve supply is from the first and second sacral nerves within the pelvis and, therefore, would not be encountered at total hip arthroplasty. The pyriformis may be partly blended with the gluteus medius. It may also be absent.

Figure 1.34. Orientation of the fibers of the hip capsule.

The capsule of the hip lies deep to the gluteus minimus (Figure 1.34). There are, however, capsular fibers of the gluteus minimus located superiorly and of the iliacus located anteriorly and medially. Once the capsule has been exposed, the reflected head of the rectus femoris will be encountered anteriorly and superiorly. This is a tough, fibrous, tendinous insertion that reinforces the superior portion of the hip capsule as it inserts into the superior margin of the acetabulum. The direct head of the rectus femoris inserts onto the anterior-inferior spine, which is also a landmark for exposure. Exposure of the acetabulum is usually facilitated by passing a retractor beneath the reflected head and over the iliopectineal eminence of the superior ramus of the pubis. The innervation of the rectus femoris is from the femoral nerve and well out of the area of exposure for total hip arthroplasty. However, the femoral nerve runs over the superior ramus of the pubis and can be damaged by sharp retractors, which are inserted over the pubis ramus into the pelvis. The other structure that passes over the brim of the pelvis just medial to the anterior-inferior iliac spine is the tendon of the psoas major, which at the level of the pelvic brim is surrounded by muscle fibers of the iliacus. The psoas tendon inserts into the lesser trochanter of the femur. The muscle fibers of the iliacus extend distal to the lesser trochanter to insert onto the body of the femur in front of and below the lesser trochanter. There is usually an indentation in the anterior lip of the acetabulum where the psoas crosses it. The psoas serves to reinforce the Y ligament of Bigelow as the hip is extended. The combination forces an anteverted head into internal rotation, which accounts for some of the abnormalities of gait associated with femoral anteversion. It is important when passing retractors over the anterior acetabular 1ip and the superior ramus of the pubis not to damage the psoas tendon. It is also important to make sure that no cement gets between the pelvis and the psoas because an inflammatory and mechanical irritation can be created. The position of the iliacus muscle fibers and psoas tendon lateral to the femoral vessels and nerve take on greater significance in the case of fixed flexion contracture, where psoas release may be necessary. The psoas must be separated from the inferior capsule in cases where an inferior capsulectomy may be necessary in a very stiff hip. If lengthening of the psoas is necessary, we prefer to identify the tendon at the brim of the pelvis and separate only the tendinous portion so that an effective lengthening can be achieved without running the risk of producing a complete discontinuity. The fibers of the iliacus will allow the muscle to stay in continuity with its insertion, allowing lengthening without complete transection.

Figure 1.35. Details of the short external rotators.

The External Rotators of the Hip

When carrying out total hip arthroplasties through the posterior approach, the external rotators are transected and should be reattached (Figure 1.35). The pyriformis has already been described; in addition to being an abductor, it is also an external rotator. The obturator internus and gemelli form a common insertion just inside the tip of the trochanter and deep to the pyriformis tendon.The obturator internus originates from the inside of the obturator foramen, passing out of the pelvis through the lesser sciatic foramen and then passing horizontally across the posterior capsule of the hip, where it receives the attachments of the gemelli and is inserted into the aforementioned spot on the trochanter. Its innervation comes from a special nerve from the sacral plexus within the pelvis.

The obturator externus covers the outer surface of the anterior wall of the pelvis, arising from the margin of the medial side of the obturator foramen. The fibers end in a tendon that runs across the back of the neck of the femur and inserts into the trochanteric fossa. It is innervated from a branch of the obturator nerve. The last of the important short external rotators is the quadratus femoris, which arises from the upper part of the external border of the tuberosity of the ilium and inserts into the upper part of the linea quadrata extending downward from the intertrochanteric crest. Superior to the quadratus femoris is the gemellus inferior, and inferior to it is the adductor magnus. It is innervated from a branch from the sacral plexus. The quadratus femoris marks the inferior margin of the muscle release necessary for exposure of the hip through the posterior approach.

The sciatic nerve lies deep to the pyriformis muscle but superficial to the rest of the external rotators. It passes sufficiently medial to the insertion of the external rotators and to the posterior aspect of the femur that it is seldom at jeopardy during total hip arthroplasty. If there is a great deal of abnormality requiring reconstruction of the posterior wall of the acetabulum, then its position needs to be specifically determined so that it can be protected from damage by the retractors.

Vessels About the Hip

The lateral, anterolateral, and posterior approaches to the hip do not generally encounter any significant vascular problems. The common iliac artery and vein lie on the anterior surface of the wing of the ilium and cross the superior pubic ramus and pass medial to the femoral head. Although they are generally not visualized at total hip arthroplasty, they could be damaged by sharp retractors inappropriately placed. The medial femoral circumflex artery arises from the medial aspect of the profundus and passes between the pectineus and the psoas major. Dissection in this area can result in brisk bleeding from one of its major branches. The acetabular branch from the medial femoral circumflex enters the hip joint beneath the transverse ligament and supplies blood to the fat in the bottom of the acetabular fossa. Sharp dissection in this area can also encounter brisk bleeding. The lateral circumflex artery arises from the lateral side of the profunda and passes behind the rectus femoris, dividing into anterior, transverse, and descending branches. The terminal divisions of the transverse branch wind around the femur just below the greater trochanter and may be encountered when splitting the vastus lateralis fibers in carrying out the direct lateral approach. The superior gluteal artery passes out of the greater sciatic notch above the pyriformis in the company of the superior gluteal nerve and passes between the medius and minimus. The inferior gluteal artery comes out below the pyriformis and has arterial branches that overlie the short rotators. Most of these vessels are terminal branches of relatively minor arteries, and although they can be annoying at total hip arthroplasty, particularly when a great deal of dissection is necessary to provide mobility, they are seldom of great consequence.


This brief review of anatomical and mechanical factors important to total hip arthroplasty provides a perspective for our approach to the problems of this frequent procedure. Understanding of the forces that cross the hip and of the details of the anatomy leads to a better understanding of some of the failures of the past and gives credence to current solutions.


  1. Charnley J Sr. Long term results of low friction arthroplasty. In: Proceedings of the 10th Open Scientific Meeting of The Hip Society CV Mosby, St. Louis, pp 42-49, 1982
  2. Getz B. The hip joint in Lapps. Acta Orthoi Scand (Suppl 22) 1955
  3. Hoagland FT, Low WD. Anatomy of the femoral neck and head, with comparative data from Caucasians and Hong Kong Chinese. Clin Orthop 152:10, 1980
  4. Kapandji Al. La Physiologie Articulaire, vol. 2. Libraire Malone, Paris, 1980
  5. Kingsley PC, Olmsted KL. A study to determine the angle of anteversion of the neck of the femur. J Bone Joint Surg (Am) 30:745, 1948
  6. McKibbin B. Anatomical factors of the hip joint in the newborn. J Bone Joint Surg (Br 52:148, 1970
  7. Pauwels F. Biomechanics of the Locomotor Apparatus. Springer Verlag, New York, pp 1-228, 1980
  8. Radin EL. Biomechanics of the human hip Clin Orthop 152:28, 1980
  9. Steindler A. Mechanics of Normal and Pathological Locomotion in Man. Charles C Thomas, Springfield IL, 1935
  10. Vasu R, Carter DR, Harris WH. Stress distributions in the acetabular region. I. Before and after total joint replacement. I Biomech 15:155, 1982
  11. von Lanz T. Anatomische und entwicklungsgeschichtliche Probleme am Huftgelenk. Verhandlungen der Deutschen Orthopadischen Gesellshaft 37:7, 1949