Javascript not enabled
ResearchFurther Opinion

Anatomy of piriformis, obturator internus and obturator externus

LB Solomon, YC Lee, SA Callary, M Beck, DW Howie

J Bone Joint Surg [Br] 2010;92-B:1317-24.


The authors report a rather elegant study whose aim was to investigate the anatomy of the piriformis and the two obturator muscles as they relate to the posterior approach in total hip arthroplasty.  There were two distinct parts to their work.  The first was based on the dissection of ten embalmed cadavers and the second by representing each of the three muscles by a series of strings on a model pelvis and femur.

In the cadaver study each of the paired hips were dissected until only gluteus medius, gluteus minimus, piriformis, obturators internus and externus and the bony ligamentous and capsular structures remained on the pelvis.  The origins and insertions of the piriformis, obturator internus and obturator externus and their relationships to neighbouring structures were then recorded and their lengths measured using a caliper. 

In contrast to traditional reporting of a separate insertion of piriformis and obturator internus, the authors’ dissections demonstrated that in every case the cylindrical tendons of piriformis and obturator internus joined to form a conjoint tendon before inserting on to the proximal femur.  The length of this conjoint tendon varied from 0.5 to 2 cm and it flattened before inserting on to the superior margin and the entire medial surface of the greater trochanter all the way to the proximal part of the anterior intertrochanteric line.  The conjoint tendon had connections to the joint capsule, the posterior margin of gluteus medius and obturator externus. Also as emphasised by the authors anterior to the connection between the conjoint tendon and gluteus medius they were separated by a fat pad with the gluteus medius taking a lateral insertion on the greater trochanter.  Apart from the fibrous connection with the hip joint capsule the conjoint tendon was easily separated medially and inferiorly from the femoral neck and hip capsule.  The authors also commented that obturator externus had very few expansions into the joint capsule which meant that if it is not identified and tied at the time of exposure then once cut it will retract and will not be repairable.

The authors noted that the complete division of the connection between the conjoint tendon and the gluteus medius allowed the superior margin of the greater trochanter to be uncovered with ease from the deep surface of gluteus medius and thus the entire medial surface of the greater trochanter could be exposed without having to retract gluteus medius.  This particular dissection was done on each right hemi pelvis. The authors then say that they observed the degree of retraction of gluteus medius that was necessary to expose the medial aspect of the trochanter as compared to the left hemi pelvis where piriformis and obturator internus were simply detached. However the authors didn’t give any objective evaluation of this.

The other part of the study was based on a saw bone model of pelvis and femur.  Each muscle was represented by a series of strings, three for piriformis, six for obturator internus and five for obturator externus.  Each string was fixed at the femoral insertion of the muscle and then passed through an eyelet at the site of origin of the muscle on the pelvis and then on to a wooden board.  As the model femur was moved then the change in length of each muscle segment was measured.  Then using a commercially available navigation system the femur was moved relative to the pelvis so that it was positioned as for preparation of a femur during total hip replacement using the posterior approach.  This positioning was again based on information gathered from intra-operative navigation data.  The position involved rotating the femur such that the lower leg was vertical whereas many surgeons actually rotate the femur beyond this position.  So in other words the position they adopted represented a conservative movement of the femur relative to the pelvis. 

This showed that the maximum lengthening for piriformis was 182%, 185% for obturator internus and up to 220% for obturator externus.  According to references quoted by the authors1,2 when a muscle is stretched beyond approximately 180% of its length it is damaged beyond repair and thus the authors concluded that failure to detach these muscles at the time of surgery may result in partial or complete irreversible damage.  In terms of preserving piriformis the authors suggest an alternative which is division of the conjoint tendon which joins obturator internus to piriformis and then releasing the piriformis from the trochanter but maintaining its connection to gluteus medius.  This particular dissection was not done as part of their study.

The authors comment that there has been no objective evidence to support less invasive procedures which preserve either piriformis3-5 or obturator externus muscles6 objective investigations using gate analysis have been unable to demonstrate early improved function after total hip replacement performed through minimal invasive surgery compared with classic techniques.7

This paper certainly questions the rationale of preserving piriformis and or obturator externus during the posterior approach in total hip replacement. For those surgeons who preserve piriformis the authors’ suggestion of maintaining its attachment to gluteus medius would appear attractive.


1. Gordon A, Huxley A, Julian F. The length-tension diagram of single vertebrate striated muscle fibres. J Physiol 1964;171:28-30.
2. Guyton C, Hall J. Textbook of Medical Physiology. Tenth ed. W.B. Saunders, 2000.
3. Khan RJ, Fick D, Khoo P, Yao F, Nivbrant B, Wood D. Less invasive total hip arthroplasty: description of a new technique. J Arthroplasty 2006;21:1038-46.
4. Prigent F. Incidence of capsular closure and piriformis preservation on the prevention of dislocation after total hip arthroplasty through the minimal posterior approach: comparative series of 196 patients. Eur J Orthop Surg Traumatol 2008;18:333-7.
5. Procyk S. Initial results with a mini-posterior approach for total hip arthroplasty. Int Orthop 2007;31(Suppl):17-20.
6. Kim YS, Kwon SY, Sun DH, Han SK, Maloney WJ. Modified posterior approach to total hip arthroplasty to enhance joint stability. Clin Orthop 2008;466:294-9.
7. Ward SR, Jones RE, Long WT, Thomas DJ, Dorr LD. Functional recovery of muscles after minimally invasive total hip arthroplasty. Instr Course Lect 2008;57:249-54.


Beverland D

Musgrave Park Hospital, Belfast, United Kingdom