The hip and pelvic region was dissected in eighteen specimens harvested from fifteen embalmed human cadavers. The specimens had been obtained from eight female and seven male donors who had had a mean age of seventy-eight years (range, sixty-five to ninety-four years) at the time of death. Details regarding the antemortem hip function and mobility status of the donors were not available. Standard lateral and posteroanterior radiographs of each hip joint were made and analyzed by a radiologist to establish the presence of joint abnormalities or surgical implants.
Dissection and Injection Methods
Dissection was undertaken in stages to identify the bursae beneath the layered gluteal tendons. The skin and superficial fascia covering the thigh and buttock were removed to expose the fascia lata and the gluteus maximus muscle. Beginning at the level of the knee joint, the fascia lata was reflected in a posterosuperior direction up to the point of insertion of the distal tendinous fibers of the gluteus maximus onto the femur. An incision was then made along the iliac crest, through the layer of deep fascia covering the gluteus medius muscle, and, posteriorly, through the iliac attachments of the gluteus maximus muscle. The tensor fasciae latae was released from its proximal insertions, was separated from the gluteus medius muscle, and, together, with the fascia lata, was reflected posteriorly until the edge of a bursa was exposed. In the majority of cases, bursal tissue was easily identified as it contrasted in texture and hue from the surrounding fatty connective tissue, with each bursa having a definite outline and shape. The outer surfaces of some bursae were peppered with small blood vessels, and most contained a thin film of fluid, which aided identification16. Once a bursa had been identified, the surrounding tissues were cleared away. Reflection of the fascia lata continued until all bursae in this tissue plane had been revealed.
To improve tissue differentiation, 1 to 5 mL of moulding latex, colored with blue paint tint, was injected into each bursa16 with use of a 10-mL Luer slip syringe and a 20-gauge (0.9 × 25-mm) needle. A small amount of 1% hydrochloric acid was used to seal the opening left by the needle. One or two bursae were injected at any one time, and the latex was left to solidify overnight before proceeding further.
Once all data had been collected for the bursae deep to the muscle and tendon of the gluteus maximus and the fascia lata, further dissection was undertaken to expose the bursa or bursae beneath the gluteus medius tendon. Beginning posteriorly, the fascicles of the gluteus medius muscle were cleaved from their proximal osseous attachments and were reflected in an anteroinferior direction. The tendon of the gluteus medius was then reflected toward the greater trochanter. The first bursa to be viewed was associated with the distal tendon of the piriformis muscle. Its superficial surface frequently adhered to the deep surface of the gluteus medius tendon and, following the injection of latex, these two structures were carefully separated. The thickened, posterior part of the gluteus medius tendon was then incised and was peeled in an anteroposterior direction to reveal the bursa or bursae that were positioned anteriorly. Once the bursa or bursae had been injected and separated from the gluteus medius, the tendon was reflected further.
Reflection of the gluteus medius allowed access to the gluteus minimus, which was cleaved from its proximal iliac insertions and was reflected inferiorly. In order to expose the bursa or bursae, the distal tendon of the gluteus minimus had to be cut from its insertions into the hip joint capsule and the greater trochanter. Specifically, the posterior portion of the tendon was incised and was reflected from its insertion onto the upper border of the trochanter, allowing the anterior part of the tendon to be reflected in an anteroinferior direction. The bursae deep to the tendon of the gluteus minimus were then injected with latex.
Measurements
For each bursa, the points of greatest height and width were measured with use of calipers with an accuracy of ±1 mm (EBM; Gneunel, Switzerland) and the basic shape of the bursa was noted. This information was then used to calculate area according to one of two formulae. The area of circular/oval bursae was calculated as ¾ × height × width, whereas the area of triangular bursae was calculated as ½ × height × width. For each different type of bursa that was identified, the mean and the standard deviation of the mean were calculated. The location of each bursa in relation to the surrounding structures was noted and photographed, and these data were analyzed. The position of the center of each bursa was measured relative to the superior tip (apex) of the greater trochanter.
Histological Analysis
A small sample of tissue was removed from each identified bursa, was post-fixed in 10% neutral buffered formalin, and was processed for light microscopy. Five-micrometer sections were cut and were stained with hematoxylin and eosin. Two investigators (S.J.W. and H.D.N.) independently performed histological analysis to establish that each section was composed of bursal tissue, as determined by the presence of a distinct synovial lining20,24,25. Images of each section were captured with use of a SPOT RT color digital camera (software version 3.5.4; Diagnostic Instruments, Sterling Heights, Michigan). The digital images of samples from each bursa were further analyzed, and the synovium was classified with use of an adaptation of the guidelines described by Key25.
In the eighteen specimens, a total of 106 bursae were identified beneath the distal tendons of the three gluteal muscles, representing ten distinct bursae. Schematics of these bursae, showing their location relative to the apex of the greater trochanter and the gluteal tendons, are provided in Figures 1-A, 1-B, and 1-C. The mean number of bursae per hip was six, with a minimum of four and a maximum of nine bursae being found in any one specimen. Five other structures were injected with latex, but, because of the absence of a synovial lining, they were not verified as being bursae. Radiographically, no surgical implants were present, and mild to moderate osteoarthritic changes of the hip joint were detected in eight of the eighteen specimens.
Bursae Deep to the Gluteus Maximus Muscle and Tendon and the Fascia Lata
On the basis of location and size, four different bursae were located deep to the gluteus maximus and the fascia lata, with at least one bursa being present in each specimen, giving a total of forty-eight bursae. Three distinct bursae were identified in the area immediately lateral to the greater trochanter and were identified, according to the terminology used by Dunn et al.16, as the deep, secondary deep, and superficial subgluteus maximus bursae. The fourth structure, the gluteofemoral bursa, was more distal in location (Figs. 1-A and 2). The dimensions and locations of these bursae are summarized in Tables I and II.
The dominant, deep subgluteus maximus bursa is normally considered the "trochanteric bursa" and was present in sixteen of the eighteen specimens. Lying deep to the fascia lata and/or the distal fibers of the gluteus maximus muscle, this large round or oval bursa was centered over the lateral surface of the greater trochanter. It was separated from the bone only by the distal tendon of gluteus medius and the origins of vastus lateralis, which it partially covered (Figs. 1-A and 2). No other "trochanteric" bursae were identified in four of these sixteen specimens but, in these four specimens, a gluteofemoral bursa was always present.
The most common pattern (observed in eight of sixteen specimens) was the presence of the deep subgluteus maximus bursa together with the superficial subgluteus maximus bursa (Figs. 1-A and 2). The oval superficial bursa was superficial (lateral) to the deep bursa, was located beneath the distal fibers of the gluteus maximus muscle and/or the fascia lata, and was often attached to, and reflected with, these tissues. The location of its center point appeared to be similar to that of the deep bursa.
In six specimens, the secondary deep subgluteus maximus bursa was present in the same plane as the dominant deep subgluteus maximus bursa (Fig. 1-A). In two specimens, communication was evident between the opposing walls of these two deep bursae. The secondary deep subgluteus maximus bursa was posterior to the dominant deep bursa and in most cases covered part of the posterior border of the greater trochanter.
The gluteofemoral bursa was identified in all specimens but one, and in two specimens it was the only bursa that was present in the plane deep to the gluteus maximus muscle. Ten of the gluteofemoral bursae were deemed to be associated with the greater trochanter as the superior portion of each bursa covered the lower third of this landmark (Figs. 1-A and 2). These bursae were relatively large and were either triangular or pear-shaped. The distal end or posterior border of the gluteofemoral bursa associated with the trochanter was positioned beneath, and adhered to, the iliotibial band in the area where the tendinous fibers of gluteus maximus inserted. Because of their smaller size (Table I), the remaining seven gluteofemoral bursae in these seventeen specimens had no association with the trochanter. These bursae were positioned over the posterior edge of the vastus lateralis, separating it from the iliotibial band, and they also covered some of the proximal tendon of the vastus lateralis.
Bursae Deep to the Tendon of the Gluteus Medius Muscle
Thirty-six bursae were located deep to the tendon of gluteus medius, and all were associated with the anterosuperior portion of the greater trochanter. These bursae were found in two different areas and, to reflect their location, they were named the anterior subgluteus medius bursa and the piriformis (posterior subgluteus medius) bursa (Figs. 1-B and 3). Of the eighteen specimens, four had a single bursa, eleven had two bursae, two had three bursae, and one had four bursae. These bursae were considerably smaller than the bursae found deep to gluteus maximus (Table I).
Sixteen anterior subgluteus medius bursae, located deep to the gluteus medius tendon and lying anterior to both the piriformis bursa and the apex of the greater trochanter, were identified in twelve specimens. These bursae were beneath, and usually adhered to, the deep tendon of the gluteus medius, separating the tendon from either the distal insertion of gluteus minimus and/or the lateral anterior surface of the trochanter (Figs. 1-B and 3).
The piriformis (posterior subgluteus medius) bursa, associated with the tendon of the piriformis muscle at its insertion onto the greater trochanter, was identified in sixteen of the eighteen specimens (Figs. 1-B and 3). Typically, a single round or oval bursa was present in this area, but in four specimens an additional bursa (the secondary piriformis bursa) was found (Fig. 1-B). The piriformis bursa was close to the tip of the trochanter (Table II), and its superficial surface was in contact with, and often adhered to, the deep surface of the gluteus medius tendon. In most cases, the posterosuperior portion of the bursa was positioned over the posterior insertional fibers of the piriformis and the bursa lay on and followed the contour of the tendon. Its inferior portion came into contact with the greater trochanter, covering part of its superolateral border. The secondary piriformis bursa was similar in shape and size to the piriformis bursa but differed in location, being slightly anterior to the tip of the greater trochanter (Tables I and II).
Bursae Deep to the Tendon of the Gluteus Minimus Muscle
Twenty-two bursae in two different locations were identified deep to the tendon of the gluteus minimus. The oval or round subgluteus minimus bursa was the most proximal of these bursae and was identified in fifteen of the eighteen specimens (Figs. 1-C and 4). It was situated deep to the anterior border of the tendon of the gluteus minimus as it passed around the anterior aspect of the greater trochanter to its insertion. Located approximately 3 to 4 cm superior to the most distal insertional fibers of the gluteus minimus tendon, the subgluteus minimus bursa was also directly distal to the inferior insertion of the hip joint capsule into the base of the neck of the femur, covering the femoral tubercle26. The center of this bursa was the most anterior of all of the bursae associated with the greater trochanter (Table II), and this synovial structure was larger than the secondary subgluteus minimus bursa and all of the bursae deep to the gluteus medius (Table I).
The secondary subgluteus minimus bursa was present in seven of the eighteen specimens (Figs. 1-C and 4), and in five of these seven specimens it was found in conjunction with the subgluteus minimus bursa. The bursa was directly under and firmly adhered to the tendinous fibers of the gluteus minimus as they inserted into the anterolateral aspect of the greater trochanter.
Histological Findings
Various types of bursal synovial linings were observed, but the predominant tissue type in both the surface and the supporting layers of the synovium was (loose) areolar tissue (Fig. 5). Large variation was apparent in the depth of the synovial surface layer, but in general this layer appeared thicker in bursae that were located beneath the gluteus maximus and the fascia lata. Of note, the piriformis and secondary piriformis bursae differed slightly from all other bursae in that the tissue was dense areolar or fibrous in nature in thirteen of twenty specimens.
The present study demonstrated that the morphology of the bursae in the region of the greater trochanter is generally more complex than has been depicted in both the anatomical literature and clinical studies. In contrast with many standard descriptions17-21, we found that two or more bursae potentially may be present beneath each of the gluteal tendons. Notably, the findings of the present study are contrary to the long-held belief that a single "trochanteric bursa" lies over the lateral aspect of the greater trochanter, deep to the gluteus maximus and the fascia lata17,19-22,27,28. Consistent with the report by Dunn et al.16, the deep subgluteus maximus bursa typically was accompanied by either the superficial and/or secondary deep subgluteus maximus bursa or bursae and a gluteofemoral bursa. As the subgluteus maximus bursa is considered to be most culpable in the pathogenesis of so-called trochanteric bursitis3,29-33, when a bursa becomes inflamed, not just one but as many as four bursae in this area may need to be considered as a possible underlying cause of symptoms.
The treatment of refractory trochanteric pain, which could also occur as a result of other abnormalities such as abductor tendinopathy or muscle weakness12,13,15, may include the injection of corticosteroid and/or local anesthetic. However, rather than targeting a particular bursa, it appears that the administration of injectate is generally performed blindly, delivered at the site of maximum tenderness2,6,33-35. Cohen36 showed the difficulty of injecting any one of the bursae in the region of the greater trochanter, even with the aid of fluoroscopy. Therefore, it seems that having the ability to differentiate between bursae in this area would be advantageous for improving the accuracy of clinical techniques such as these.
To our knowledge, the importance of the gluteofemoral bursa as a cause of lateral hip pain has not previously been presented in the literature, possibly because no previous investigations have systematically examined this entity. This bursa has only been mentioned in passing in two studies16,29, and it is uncommon for it to be specifically named in anatomical texts19,20. Two distinct types of gluteofemoral bursae were identified in the present study and, given the proximity of the larger bursa to both the greater trochanter and the subgluteus maximus bursa, we believe that this structure warrants inclusion in future clinical and biomechanical studies of lateral hip pain.
Our findings concurred with those of Piersol23 and Hollinshead22 in that we consistently identified two or more bursae deep to the distal tendon of the gluteus medius muscle. This finding is in contrast with many other reports of a single subgluteus medius bursa14,15,17,18,20,21,26,37-39 and, on the basis of the available descriptions, it is probable that the anterior subgluteus medius bursa described in the present study is the structure that traditionally has been referred to as the solitary bursa beneath the gluteus medius tendon. Differentiating abnormalities of the subgluteus medius and subgluteus minimus bursae from their respective overlying tendons may be difficult on imaging studies15, which again emphasizes the importance of understanding the bursal anatomy in order to help to localize changes. Of note, the piriformis bursa is distinct with regard to location as it is the only bursa situated posterior to the apex of the greater trochanter, other than the secondary deep subgluteus maximus bursa. A feature unique to the subgluteus minimus and secondary subgluteus minimus bursae is their position over the anterior, rather than lateral, aspect of the trochanter. In agreement with another description17, it was also evident that the subgluteus minimus bursa was closely associated with the femoral tubercle26, which could prove to be an important radiographic marker for locating this bursa. At the present time, little is known about the role of these structures in the pathogenesis of lateral hip pain14,15, indicating the need for further research in this area.
We anticipated that a histological analysis of the bursae would help to elucidate the morphology of these structures. With few exceptions, all bursae in the greater trochanteric region had a synovial lining, but variation was evident with respect to the consistency and thickness of this layer. This finding is in agreement with those reported by Dunn et al.16, who appear to have been the only other authors to have addressed the histological characteristics of bursae in this area, specifically those beneath the gluteus maximus.
The present study has some limitations that require consideration. The cadaveric specimens were from elderly individuals, and in the future it would be useful to examine tissue from younger, living subjects. The apex of the greater trochanter was chosen to serve as a landmark from which to determine the location of the bursae, but the accuracy of palpating this point was not investigated. Also, the histological study was limited somewhat by the small numbers of bursae included in each category (with respect to tissue type and specific histological characteristics), and this precluded statistical analysis to determine if differences existed between the various types of bursae.
In conclusion, we have demonstrated that multiple bursae are associated with the greater trochanter, arranged in three distinct layers as determined by the planes of the gluteal tendons. Bursae in each layer differ with respect to morphology and location, although the centers of all bursae are positioned inferior to the apex of the greater trochanter. With the exception of the secondary deep subgluteus maximus and piriformis bursae, all are situated anterior to the apex of the greater trochanter. The multiplicity of bursae needs to be considered when assessing or managing patients who have lateral hip pain as an average of six bursae were found in our specimens in this region. The most commonly observed bursae in the vicinity of the greater trochanter were the gluteofemoral, the deep subgluteus maximus, the piriformis, the anterior subgluteus medius, the subgluteus minimus, and one of the remaining four bursae not already mentioned (the superficial subgluteus maximus, the secondary deep subgluteus maximus, the secondary piriformis, or the secondary subgluteus minimus).