Introduction

In 1995, Ganz first described femoroacetabular impingement (FAI), and, in 1999, it was introduced into the English literature. Although the first case was pincer impingement as a result of acetabular overcoverage after a periacetabular osteotomy, further investigations by Professor Ganz and his group led to the recognition of a second type of impingement: cam impingement. This type of impingement is the result of loss of the femoral head–neck offset, which causes the joint to function like a mechanical cam device. The anatomy of cam impingement has been previously described as a pistol grip deformity or a tilt deformity (Figure 19-1). Many investigators have attributed this anatomy to subclinical slipped capital femoral epiphysis (SCFE), and certainly the residuals of an unreduced SCFE can result in cam impingement. Beaule and colleagues demonstrated in a computed-tomography–based study that cam impingement can occur without SCFE, which suggests that this condition is not the result of a subclinical SCFE. Wagner and colleagues demonstrated that the bone of the cam lesion is not reactive nor does it involve any inflammation; thus it is not likely the result of the impingement or arthritis. Because the deformity occurs at or near the femoral head physeal scar, it may be genetically predetermined or the result of stresses applied during development. Either way, this pistol grip deformity has been associated with premature osteoarthritis of the hip. It has also been suggested that cam impingement results in premature or idiopathic arthritis of the hip. However, Bardakos and Villar have shown that only two thirds of patients with cam impingement show radiographic progression of arthritis at 10-year follow up.

Figure 19–1 Cam impingement. A, Anteroposterior hip radiograph (cropped from an anteroposterior pelvic view) and B, lateral radiograph of a 19-year-old collegiate basketball player with cam-type impingement with a mild crossing sign. Notice the loss of offset at the femoral head–neck junction and the convexity of the anterior femoral neck.

The anatomy of cam impingement can frequently be seen among those patients with idiopathic arthritis, although this is still a controversial subject. The suggestion that impingement results in hip arthritis is also likely when studying the pathologic findings of symptomatic impingement patients without arthritis, particularly of those with labral tears and chondral lesions that are thought to progress to arthritis when untreated. The majority of patients in Ganz’s series had a combination of cam and pincer impingement, which has been confirmed in other published series as well as my own experience of several hundred patients with symptomatic FAI. Beck and colleagues found that the demographics and pathology do correlate with the different subtypes of impingements. Particularly, for those with isolated types of impingements, the cam type is the most common (17%). It tends to occur in 19-year-old males, whereas those with isolated pincer impingement are 40-year-old active females. For those with isolated cam impingement, the common pathologic findings initially are focal, deep chondral delamination lesions (anterolaterally and extending about 1 cm from the acetabular edge) (Figure 19-2). At first the labrum is intact, but it eventually separates from the acetabular articular cartilage edge before degenerating. The labrum often separates from the acetabulum and the articular cartilage, and the articular cartilage delaminates from the bony acetabulum (Figure 19-3). Alternatively, those with pincer impingement tend to have intrasubstance crushing of the labrum, and the articular cartilage damage extends only a couple of millimeters from the acetabular edge (Figure 19-4). Although the greatest depth of penetration of articular cartilage damage in pincer impingement is also anterolateral, the damage tends to be more global and to extend around the circumference of the acetabulum. In addition, there is often posterior acetabular (62%) and femoral head (31%) articular cartilage damage as a result of the contrecoup phenomenon of the femoral head levering against the anterior acetabulum as the patient tries to obtain hip motion, which results in shearing forces posteriorly (see Figure 19-4). It has been my experience that one type of impingement will predominate the intra-articular pathologic findings.

Figure 19–2 Arthroscopic view of the chondral delamination of the acetabulum associated with cam impingement.

Figure 19–3 A, Schematic representation of the pathophysiology of joint damage with cam impingement. The cam lesion slides underneath the labrum, thereby resulting in a labral chondral separation and the detachment of the labrum from the underlying acetabulum. This occurs anterolaterally. The articular cartilage is then abutted, which results in the delamination of the cartilage from the underlying bone. B, A close-up of the region of interest where the impringement occurs.

Figure 19–4 Schematic representation of the pathophysiology of joint damage as a result of pincer impingement. The overcoverage of the acetabulum results in the crushing of the labrum against the femoral neck. This may occasionally result in the notching of the femoral neck. There is some articular cartilage damage, although this is shallower than that seen with cam impingement. With continued motion, the femoral neck levers against the acetabulum. This results in posteroinferior subluxation of the femoral head and causes posterior femoral head and acetabular articular cartilage damage.

Although the overall goal of restoring the femoral head–neck offset to relieve the abutment is the same, there is controversy with regard to the approach to be taken to address the cam lesion. Beck and colleagues found that the restoration of the femoral head–neck complex can be most reliably performed through an open approach with a trochanteric osteotomy and surgical dislocation. Although the open approach to this problem is beyond the scope of this chapter, Ganz and colleagues stated that the hip arthroscopy “technique is difficult. Simultaneous assessment of movement of the hip and debridement is not possible.” Furthermore, it has been suggested that the open approach, which is the gold standard, allows one to see the entire femoral head and the head–neck junction and thus allows for the use of templates to standardize the resection. This, in combination with the potential for complications of hip arthroscopy such as “nerve traction palsies, foot or perineal pressure sores, and iatrogenic damage to the articular cartilage of the joint,” has led many to believe that arthroscopy is limited with regard to its usefulness. However, advances in hip arthroscopy and particularly in peripheral compartment arthroscopy without traction have enhanced the ability to arthroscopically observe impingement dynamically during arthroscopic osteoplasty of the femoral head–neck junction. It has also been suggested that the “constrained hip renders access to the underlying cause of impingement technically challenging, if not impossible.” However, Sussman and colleagues performed a cadaveric comparison of open and arthroscopic techniques for cam impingement. With the use of subtraction computed tomography, those authors demonstrated that the accuracy and precision of arthroscopic osteoplasty approach that of open osteoplasty. They did report that the time for resection was faster with the open technique, although the time required to create the approach (e.g., trochanteric osteotomy, surgical dislocation) was not included. Those who perform the femoral osteoplasty or cheilectomy arthroscopically prefer this technique, because open surgical dislocation involves prolonged postoperative hospitalization (up to a week as compared with outpatient arthroscopic surgery), significant blood loss, a risk of trochanteric nonunion, and prolonged limited weight bearing (up to 12 weeks of crutch use as compared with 0 to 6 weeks). In addition, it is quite easy to see the entire central compartment arthroscopically as compared with the open surgical dislocation.

Indications

Although some believe that FAI results in arthritis and thus that surgery should be performed to prevent the arthritis, that has not been the approach used in our practice, because there is no evidence at this point that arthritis can be prevented. I certainly believe that having the anatomy of impingement does put the patient at risk for chondral injury, labral injury, and, potentially, arthritis. However, on the basis of my experience with patients who are more than 60 to 70 years old with the anatomy of FAI but no evidence of arthritis or hip symptoms in combination with my extensive experience of cadaveric research involving specimens 80 to 90 years old with the anatomy of obvious cam and combined impingement without evidence of arthritis, I have concluded that not everyone with the anatomy of FAI will develop osteoarthritis (Figure 19-5). It is my belief that the anatomy of FAI does put patients at potential risk for joint damage. However, it likely requires the individual to be involved in activities that require greater hip range of motion with or without pivoting (e.g., martial arts, soccer, running, golf) to result in impingement. After the tissues start breaking down, patients develop symptoms, because the labrum is a structure that is richly innervated. It is as soon as patients have confirmed intra-articular pain that surgery is indicated. Intra-articular anesthetic guided by fluoroscopy or ultrasound and that is given by itself or with contrast when performing magnetic resonance arthrography is a useful diagnostic test to confirm that the joint is the source of pain. Thus, the goal of surgery is to relieve intra-articular hip pain that is the result of impingement.

Figure 19–5 Photograph of a cadaveric specimen from an 85-year-old patient. Notice the obvious cam lesion and the lack of arthritic change.

Jäger and colleagues demonstrated that there is no role for physical therapy in the treatment of FAI. Their finding is not unexpected, because the bony problem will not resolve with physical therapy. Furthermore, the structures that are injured have limited if any capacity to heal spontaneously. That being said, conservative management consists of activity modification (particularly the avoidance of extremes of motion and of flexion and internal rotation particularly) and nonsteroidal anti-inflammatory medications.

The pathologies that need to be addressed include labral pathology and chondral lesions in addition to the underlying bony cause. Nearly 90% of patients who undergo hip arthroscopy for labral tears have associated bony pathology. It has been shown that those undergoing hip arthroscopy for labral tears in which the FAI was not addressed had poorer results than those without FAI. Thus, addressing intra-articular pathology without addressing the underlying cause will be less likely to result in a good outcome. Although magnetic resonance imaging is not as good as one would hope for identifying chondral damage, Johnston and colleagues found that those patients with symptomatic cam impingement and an alpha angle of more than 62 degrees are at increased risk for chondral injury. As a result, all patients undergoing surgery for intra-articular damage (e.g., symptomatic labral tears, chondral lesions) who have the anatomy of FAI should have the bony impingement treated at the same time.

Thus, indications for surgery include the following:

History and physical examination

Those patients with FAI generally describe an insidious onset of groin aching or pain. Although the condition is frequently confused with hip tendonitis or other problems, patients will often note difficulty putting on or taking off their socks and shoes. The pain is usually described as being in the groin, the inguinal region, or deep inside the joint. The pain may be worsened with activities, particularly running and other impact types of activities. Sitting (especially in low seats or chairs) for prolonged periods of time may also result in pain, and there is frequently pain when arising from a seated position. Patients with cam impingement may also have pain when squatting, cutting, or pivoting or when making sudden stops and starts. Stair climbing may also be problematic for patients with impingement. Patients may note the limited hip range of motion, particularly during flexion, adduction, and internal rotation. If the patient has an associated labral tear or a chondral flap, there may be an acute onset of symptoms as well as mechanical symptoms (e.g., locking, catching). It is not uncommon for patients to complain of hip or groin pain for years.

It is also not uncommon for patients to have had other surgeries that may not have relieved their symptoms. This may be the result of radiating symptoms or the fact that cam impingement results in limited hip range of motion. The limited motion within the femoroacetabular joint may put stress on other structures, which may result in pain or injury in these remote locations as athletic patients try to get motion to perform their activities, such as the pubic symphysis (osteitis pubis), the sacroiliac joint (sacroiliac joint dysfunction), the lower back (strains, herniated disc), and the abdominal musculature (sports hernia/athletic pubalgia). As such, several patients have been successfully treated with arthroscopic FAI in my practice (and in other practices) who have had previous laparoscopy, laparotomy, inguinal hernia repair, athletic pubalgia surgery, osteitis pubis injections and surgery, orchiectomy, oophorectomy, lumbar spine injections, decompressions, diskectomies, and lumbar spine fusion.

Although a complete discussion of the evaluation of the hip is beyond the scope of this chapter, a brief discussion of the general concepts is appropriate. First, evaluation includes the inspection of the gait and of the skin around the hip. Patients are assessed for hip weakness and tightness with the use of the Trendelenburg and Ober tests. Hip range of motion is assessed while the patient is supine. Evaluation of hip adduction and abduction, as well as internal and external rotation, should be performed in hip flexion and extension. Also, hip motions evaluated include flexion, extension, and flexion contracture. There is usually limited hip internal rotation, particularly when the hip is in flexion, among patients with impingement. Furthermore, patients frequently have pain when the hip is flexed to 90 degrees, adducted, and internally rotated; this is known as the impingement test (Figure 19-6). The labral stress test and the resisted straight-leg raise are tests that commonly result in hip pain among patients with labral tears and symptomatic impingement (Figure 19-7). These tests are often positive among patients with both cam and pincer types of FAI, and they may also be positive among patients with other sources of intra-articular hip pain.

Figure 19–6 Impingement test. The hip is flexed to 90 degrees, adducted, and internally rotated. The test is positive if pain is elicited. However, this test is not positive only in cases of femoroacetabular impingement.

Figure 19–7 The labral stress test: A, The supine patient’s hip is brought to the position of flexion, abduction, and external rotation and B, then brought into extension, adduction, and internal rotation.

Imaging and diagnostic tests

Plain radiographs are extremely valuable for the assessment of patients with hip pain that is the result of hip impingement. The standard imaging series for patients with hip pain includes an anteroposterior pelvic view with the coccyx centered 1 cm to 3 cm above the pubic symphysis and a true cross-table lateral radiograph (Figure 19-8). A frog-leg lateral view will demonstrate a lateral projection of the proximal femur and thus can be used for cam impingement assessment; however, this is not a lateral view of the acetabulum, so it has limited usefulness (see Figure 19-8, B). A cross-table lateral view, a Dunn view, and a modified Dunn view are true lateral views of the hip that can provide more information about the acetabulum (see Figure 19-8, C). The femoral head is generally symmetric, particularly the head–neck offset. A loss of the sphericity of the femoral head–neck region may be consistent with cam impingement (see Figure 19-8, A through C). This can be seen as a flattening of the concave surface of the lateral femoral neck and the appearance that the femoral head is not centered over the femoral neck. Leunig and colleagues demonstrated that, for patients with hip dysplasia, the apex of the femoral head is approximately 1 cm beyond the low point of the femoral neck, whereas in patients with impingement this distance was only 3 mm. In some situations, there may be a bump on the anterolateral surface of the femoral neck that may project beyond the femoral head or have a sharp transition or even a hook appearance at the head–neck junction. The alpha angle was originally described by Notzli and colleagues to quantify the head–neck offset on radially generated axial magnetic resonance imaging cuts of the femoral neck and head. These authors demonstrated that, in their normal population, the alpha angle averaged 42 degrees, whereas in those patients with impingement, this angle averaged 74 degrees. Most surgeons use 50 degrees or 55 degrees as their cutoff point for defining cam impingement. This angle has also been used when evaluating plain radiographs and computed tomography scans, although it has not been validated for these modalities. Additional plain radiographs may demonstrate a short femoral neck or a femoral neck–shaft angle that is varus, which may result in cam-type impingement. Untreated or residual deformity from SCFE or Legg-Calvé-Perthes disease may be seen on plain films, and this may result in cam impingement. Pincer impingement may also be seen on plain radiographs in association with coxa profunda, protrusio, retroversion, or relative retroversion of the superior acetabulum and arthritic changes.

Figure 19–8 Plain radiographs: anteroposterior pelvic and lateral views of a 22-year-old professional football player with cam impingement. A, Anteroposterior radiograph of the pelvis that demonstrates cam impingement. B, The frog-leg lateral view also shows the loss of the femoral head–neck offset. However, this view is not a lateral view of the acetabulum but only of the proximal femur. C, A true lateral view (this time of a collegiate lacrosse player). A Dunn view, or a modified Dunn lateral view may also be obtained.

Computed tomography scans, particularly three-dimensional ones, are particularly useful for demonstrating the bony anatomy associated with cam impingement (Figure 19-9). Magnetic resonance imaging, particularly magnetic resonance arthrography, is beneficial for demonstrating the cam lesion by allowing for a way to measure the alpha angle and to demonstrate labral tears, edema, or cysts within the femoral neck (these are often seen with impingement); this type of imaging can occasionally demonstrate chondral lesions (Figure 19-10). Local anesthetic is usually introduced with the contrast used for magnetic resonance arthrography to determine whether the pain is temporarily relieved within the joint, which confirms the source of pain as being intra-articular.

Figure 19–9 Representative three-dimensional computed tomography scans of 3 patients: A, a 40-year-old recreational athlete; B and C, a 35-year-old world-class triathlete; and D, a 27-year-old recreational softball player. These images demonstrate the loss of the offset at the femoral head–neck junction, thus showing the usefulness of this type of scan for the identification of the extent of the cam lesion. Figure D also demonstrates an acetabular rim fracture seen with combined cam and pincer FAI.

Figure 19–10 A, Magnetic resonance arthrogram of a classic labrochondral separation in a patient with a cam lesion. B, Magnetic resonance arthrogram demonstrating an intrasubstance labral tear in a 32-year-old male laborer with combined FAI.

Surgical technique

The goals of surgery are to relieve the abutment between the femoral head–neck junction and the acetabular rim and to treat the associated pathology (i.e., labral tears and chondral lesions; Figure 19-11, A through D). Although this may be done with the patient in the supine or lateral position, my preference is the supine position, which is described in Chapter 9 of this book.

Figure 19–11 A, Anteroposterior pelvic, and B, lateral hip radiograph of a 40-year-old engineer who had to give up martial arts as a result of hip pain. C and D, The same patient’s postoperative radiographs 6 months after resection.

Surgery for cam impingement starts with the evaluation and treatment of pathology in the central compartment. With the patient lateralized relative to the perineal post and traction applied, the three standard central compartment portals—anterior, anterolateral, and posterolateral—are made. The anterolateral and posterolateral portals are the well-described portals that are made at the anterior and posterior margins just proximal to the tip of the greater trochanter. The modified anterior portal that we have used for the past 3 years is 7 cm distal and anteromedial to the anterolateral portal at a 45-degree angle. This portal has been used for the following reasons: 1) it reduces the risk of injury to the lateral femoral cutaneous nerve; 2) it reduces the risk of postoperative rectus femoris tendonitis; and 3) it allows for a better approach to the joint if a labral repair becomes necessary. The entire joint is inspected with 70-degree and 30-degree lenses. All central compartment pathology is addressed before moving to the peripheral compartment to perform the cheilectomy or the femoral osteoplasty.

Chondral flaps are removed, and chondral defects and lesions are debrided to a stable edge. If the lesion is sufficiently large enough, a microfracture is performed (Figure 19-12, A through C). Our experience has been that femoral head chondral lesions do not do as well postoperatively as acetabular lesions do. However, when there are chondral lesions of the acetabular articular surface, the results of hip arthroscopy in general, including FAI surgery, are also less predictable as compared with patients in whom the articular cartilage is intact.

Figure 19–12 The hip of a 32-year-old competitive triathlete. A, Chondral lesion at surgery with delamination of the articular cartilage. B, After microfracture. C, A second look 7 months after the initial surgery.

Labral tears are addressed next. If the labrum is intact and there is combined impingement, then an anterior portal at the junction of the anterosuperior iliac spine and the greater trochanter is made. An arthroscopic knife is then introduced to detach the labrum at the site of acetabular overcoverage or pincer impingement. A traction suture may be used to retract the detached labrum. The acetabuloplasty may then be performed with the use of a motorized burr. After other pathology has been addressed—including synovectomy, chondroplasty, and microfracture—the labrum is reattached with the use of suture anchors (Figure 19-13). If there is a labrochondral separation, which occurs more commonly with cam impingement, and if the labrum has minimal or no obvious intrasubstance damage, then this may be repaired with the use of suture anchors. Healing occurs from the acetabular bony blood supply and capsule. If there is significant intrasubstance damage and tearing of the labrum, then a partial labrectomy is carried out; there is limited blood supply within the labrum, and, as a result, the capacity to heal is limited. Partial labrectomy may be performed with the use of meniscal-type biters, shavers, or radiofrequency devices.

Figure 19–13 A, Labral tear and B, repair in a 29-year-old competitive marathon runner.

After the central compartment pathology has been addressed, attention is turned to the peripheral compartment and the cam lesion. This is performed with the traction taken off of the patient either by releasing the traction or taking the foot out of the traction boot altogether. The preferred technique on the fracture table is to keep the foot in the foot holder of the fracture table. The traction is removed to relax the capsuloligamentous structures, thus making the peripheral compartment more capacious and allowing for easier maneuverability.

Although many surgeons prefer to perform peripheral compartment arthroscopy with the hip flexed 20 degrees to 45 degrees to further relax the anterior capsule, I prefer to perform cam surgery decompression with the hip in neutral flexion and extension. When the extremity is in neutral flexion and extension, it is easier to get a truer fluoroscopic image, because the fluoroscope beam can be made perpendicular to the axis of the femur (Figure 19-14). Because fluoroscopy beams diverge, there is inherent distortion, which would be magnified if the central fluoroscopic beam is not perpendicular to the proximal femur; this may result in inadequate or excessive bony resection. While there are benefits to peripheral compartment arthroscopy in the neutral flexion-extension position, the trade-off, which the author accepts, is that it is necessary to perform a partial capsulectomy of the anterolateral capsule to allow for adequate visualization.

Figure 19–14 Comparison drawings of the fluoroscopic beam being perpendicular to the extremity versus the extremity being flexed. A, Because only the central beam in the fluoroscopic unit is truly perpendicular, maintaining the hip in neutral flexion and extension reduces the divergence and potential disorientation associated with B, performing the procedure with the hip in flexion with the fluoroscopic beam at an angular orientation.

To perform peripheral compartment arthroscopy with the patient in the neutral flexion and extension position, the standard anterolateral portal is used along with a proximal anterolateral portal that is 3 cm to 4 cm proximal to the standard anterolateral portal (Figure 19-15). The blunt trocar and the sheath for the arthroscopic camera are introduced via the anterolateral portal to the apex of the femoral head–neck deformity. The trochar is exchanged for the arthroscope within the cannula. The scope and cannula are maintained on the capsule and bone at the apex of the cam deformity. Next, the arthroscopic motorized shaver is introduced through the proximal anterolateral portal. With the use of triangulation, the shaver is brought to the tip of the camera in its sheath (Figure 19-16). At this point, the camera and the shaver are lying on the capsule directly over the apex of the bony deformity laterally; this is seen on the fluoroscopic monitor, which is projecting an anteroposterior view of the hip (see Figure 19-16). The shaver then is used to clear some of the soft tissue over the capsule. The shaver tip soon becomes visible, and the shaver is used to make a capsulotomy of the anterolateral capsule. This is enlarged as a partial capsulectomy to a size of 1 cm to 1.5 cm in diameter (Figure 19-17). This is generally performed with the use of a 30-degree arthroscopic lens and a 5.0-mm aggressive shaver, with the fluoroscopic monitor used for guidance. A radiofrequency device may also be used to perform the partial capsulectomy. Glick has shown that larger capsulectomies of the anterolateral capsule are safe and not associated with hip instability. Furthermore, he has shown that this capsule ultimately heals.

Figure 19–15 Patient supine with the head to the left and feet to the right. Notice the anterolateral portal (with the arthroscope in the cannula) and the proximal anterolateral portal with the arthroscopic shaver introduced into the portal. The modified anterior and posterolateral portal incisions can be seen without instrumentation placed in them.

Figure 19–16 Fluoroscopic image of the trocar in the anterolateral portal and the shaver at the location of the maximal deformity.

Figure 19–17 Arthroscopic image A, of the shaver making the capsular window for cam decompression in a 20-year-old female collegiate soccer player. B, The view of the cam lesion after partial capsulectomy utilizing this approach.

After the partial capsulectomy is completed, the shaver is introduced into the peripheral compartment. A partial synovectomy is performed to allow for the adequate visualization of the anterior, lateral, and posterior femoral head, with its cam deformity; the extra-articular acetabular rim; the labrum; and the capsular reflection on the acetabulum, in addition to many other peripheral compartment structures (see Figure 19-17). The lateral synovial fold with its retinacular vessels is also visualized, as is the medial synovial fold and the zona orbicularis. The hip can be flexed, adducted, and rotated to demonstrate the impingement arthroscopically, either with the traction boot attached to the fracture table or by removing it from the traction device.

With the use of a motorized burr, the femoral head–neck offset is restored when the surgeon removes the excessive bone. Not all cam impingement patients have the same anatomy, because the bumps differ: some may be lateral, some may be anterior, and most are anterolateral. Thus, the same operation (i.e., the location and amount of bone removed) is not performed on every patient. The surgery must be tailored to the patient’s pathology. The goal is to restore the normal anatomy and offset for the patient (Figure 19-18). Bony resection and visualization may be enhanced by exchanging the scope with the shaver or burr so that the camera is introduced in the proximal anterolateral portal, whereas the burr is introduced from the standard anterolateral portal. The hip may be rotated, flexed, abducted, or adducted to help with bony resection.

Figure 19–18 Arthroscopic view A, before and B and C, after the resection of a cam lesion in a 21-year-old collegiate football player.

The ideal amount of bone to resect has not been determined. Some clinicians start their resections 7 mm to 10 mm from the labral edge and work distally. However, to restore the alpha angle, one has to remove bone up to the labral edge. However, the removal of bone to the labral edge can result in the loss of the sealing function of the labrum. In addition, as is mentioned later in this chapter, the outcomes of decompression of the cam lesion do not correlate with the ability to restore the alpha angle to less than 50 degrees.

The amount of bone removed should be individualized on the basis of the patient’s pathology. General guidelines suggest that the resection should be less than 1 cm deep, 8 mm from proximal to distal, and 15 mm from medial to lateral. Mardones and colleagues determined that the resection of more than 30% of the femoral neck width increases the risk of fracture of the femoral neck. Thus, resection is kept to less than 30% of the femoral neck width, and usually much less than that is necessary to eliminate the impingement. Sometimes the cam lesion is well circumscribed and demarcated, whereas other times it is not. Thus, fluoroscopy can help to identify the lesion and to assess how much bone is removed to avoid over-resection and increased risk of fracture. In addition, the hip may be dynamically assessed during the bony resection to ensure the adequacy of the bony resection.

Fortunately, cam lesions are almost exclusively anterior, anterolateral, or lateral. The blood supply to the femoral head, which is supplied by the posterior circumflex vessels, travels within the lateral synovial fold and the femoral neck posterolaterally, which is a safe distance from the resection (Figure 19-19). Injury to these vessels may result in avascular necrosis of the femoral head, although I am not aware of any reported cases of this condition after FAI surgery.

Figure 19–19 Drawing of the blood supply to the femoral head. Notice that the blood vessels from the lateral circumflex vessels enter the femoral head posterolaterally.

Occasionally cysts of the femoral head–neck junction are seen arthroscopically. These cysts, which were originally thought to be normal variants, are likely the result of the abutment of the femoral neck against the labrum; they are seen in a third of patients with FAI. Thus, the arthroscopic visualization and decompression of these femoral neck cysts help to confirm the location of the impingement and the adequacy of the bony resection (Figure 19-20).

Figure 19–20 Arthroscopic view of a cyst in the femoral neck A, before and B, after resection in a 30-year-old recreational athlete.

Detailed descriptions of other techniques for cam resection are beyond the scope of this chapter. However, some considerations of other techniques will be discussed. Some clinicians prefer not to perform a partial capsulectomy. As a result, some surgeons position the hip in 20 degrees to 40 degrees of flexion. There are surgeons who perform the resection with the patient’s hip in neutral rotation, and others prefer varying degrees of internal rotation with hip flexion. There are clinicians who perform a more extensive capsulotomy or capsulectomy that includes making an anterolateral longitudinal incision in the capsule initially and then making a transverse incision in the capsule at its acetabular insertion (i.e., essentially making a “T”). Some surgeons will start their hip arthroscopy in the peripheral compartment before entering the central compartment. By doing this, these surgeons may visualize the impingement first and address the cam lesion before entering the central compartment under arthroscopic visualization, which has the benefit of reducing the risk of iatrogenic labral damage. Other portals that can be used to address the peripheral compartment have also been described, including a distal anterolateral portal that is 7.5 cm distal to the anterolateral portal and an anterior portal that is one third of the distance from the anterosuperior iliac spine to the greater trochanter.

Postoperative rehabilitation

There is very little science to guide postoperative rehabilitation after femoral osteoplasty or cheilectomy. Immediate full weight bearing after these arthroscopic procedures has been advocated by a minority of surgeons who also do not recommend physical therapy postoperatively. However, despite the work of Mardones, there have been isolated anecdotal reports of hip fracture after arthroscopic femoral osteoplasty or cheilectomy. It is my contention that postoperative stiffness and prolonged pain can be avoided by appropriate rehabilitation. As a result, I recommend a guided rehabilitation protocol that is advanced as patients reach particular milestones.

I recommend weight bearing to 20 lbs of force with crutches for 2 weeks after the procedure for most patients, provided that they have adequate bone quality and no concomitant microfracture. For women who are more than 39 years old and men who are more than 49 years old, I add an extra week of limited weight bearing per decade. We encourage early range-of-motion exercises, including the use of a continuous passive motion machine and passive circumduction exercises. Weight bearing occurs with the foot flat, because this reduces the amount of force across the hip and lessens the risk of hip flexor tendinitis. Rehabilitation includes range-of-motion exercises, stretching, and hip and core strengthening. Proprioception exercises are included early during the course of rehabilitation. The second phase of rehabilitation (usually the second month through the end of the third month) includes balance progression, stationary cycling with resistance, and increasing strength training, which includes double- and single-leg knee bends, side stepping, and elliptical training. During the fourth month, plyometrics are initiated and advanced, along with agility drills, running, and side-to-side movements. The goal is to return to sports participation between 4 and 6 months after surgery.

Results and outcomes

There have been only a few case series that have reported the results of hip arthroscopy for FAI, although the number of publications is growing rapidly. In fact, the number of patients and hips with FAI that have been treated arthroscopically is more than 800, which exceeds the number addressed in the literature regarding the open treatment of FAI. Although many of the early studies report the results of cam decompression, this is because arthroscopic acetabuloplasty techniques lagged behind. Unfortunately, most studies include patients with cam, pincer, or both types of impingements, and the outcomes and results are not differentiated. As noted previously, arthroscopic labral debridement in the setting of FAI in which the FAI is not treated does not have good results. The evaluation of the results of any procedure involving the nonarthritic hip suffers from the lack of an adequate, patient-generated, validated outcomes scale or score.

Philippon has shown that an alpha angle of more than 62 degrees correlates with chondral injury at the time of arthroscopy. However, Stähelin and colleagues and Brunner and colleagues have both reported that the outcomes of FAI surgery do not correlate with the clinical outcomes. This is not surprising, because the resection of the bone from the femoral head–neck region must be at the level of the labrum with the hip in neutral flexion and extension to restore the alpha angle for an anteriorly based loss of offset. By removing this much bone, the labrum loses its sealing effect. In addition, as noted previously, not all cam lesions are the same. The alpha angle measures the loss of anterior femoral head–neck offset, but, in reality, not all cam lesions are the same: some involve a loss of lateral offset, some are anterior, and some demonstrate both conditions; thus, the alpha angle measurement does not encompass the full extent of the pathology.

Table 19-1 reports the results of the arthroscopic treatment of FAI. A common theme that correlates with the results of the open surgical treatment studies is that concomitant chondral damage—and particularly arthritis—is a poor prognostic finding, with a larger proportion of failures occurring in this group of patients.

Complications

In general, complications related to hip arthroscopy are related to traction: too much or too little, patient positioning, and fluid management. Complication rates are between 1.4% and 5.5%. Reported complications include the inability to perform the arthroscopy as a result of access issues in addition to neuropraxias of the sciatic, femoral perineal, pudendal, and lateral femoral cutaneous nerves that often resolve spontaneously. Also reported are infections, hematomas, portal bleeding, intra-articular instrument breakage, and vaginal tears and scrotal necrosis related to excessive lateral traction force. There have been several reported cases of intra-abdominal fluid extravasation, in addition to reports of avascular necrosis (not associated with FAI surgery). Heterotopic ossification has also been noted, including after FAI surgery. Labral repairs may not heal, and, in some cases, labral repairs have been associated with capsular adhesions postoperatively, which may limit motion and cause pain. Probably the most common complications that are underreported are iatrogenic articular cartilage damage and labral injury. Although they have not been reported in the literature, deep venous thrombosis and pulmonary embolism related to hip arthroscopy have been anecdotally discussed among surgeons and at meetings and presentations where case series are presented along with their complications.

Complications related to the arthroscopic treatment of FAI include femoral neck fractures related to the over-resection of the femoral neck when treating cam impingement. Mardones and colleagues noted that the risk for femoral neck fracture is increased when the resection exceeds 30% of the femoral neck; thus, keeping the resection to less than that amount may reduce the associated risk. Another risk is the under-resection of the femoral neck, which results in the incomplete reshaping of the FAI deformity. This is also likely to be under-reported, although the exact amount and location from which to remove it still have not been defined. Although the main blood supply to the femoral head ascends on the posterolateral femoral neck, thus making it susceptible to injury during femoral osteoplasty, the cam lesion tends to occur anterolaterally and thus not in the region of the blood vessels. Nonetheless, avascular necrosis after the treatment of femoroacetabular impingement (FAI) is a potential concern, although no actual cases of the condition after FAI surgery have been reported. There is a concern that the anterior hip capsulotomy or capsulectomy may simulate injury to the iliofemoral ligament and result in hip instability; however, this phenomenon has not been reported in the literature.