Lecture Notes on Psoas & Adductors
by Robert Schleip
Psoas as a medial rotator
"In thus inserting on the medial side of the femur (the rotators insert more laterally), the psoas major can function as a medial and internal rotator of the thigh, balancing external rotators."
This statement from Ida Rolf on page 170 of her book on Rolfing is in clear contrast to the majority of anatomy books, which usually claim that the psoas is an external (or lateral) rotator of the femur. Here’s a possible biomechanical explanation for Ida Rolf’s controversial statement.
The crucial point to understand is that the axis of femural rotation does not go through the middle of the femur shaft but passes medial of it. So when the lesser trochanter is pulled anteriorly this also moves the greater trochanter forward (not backwards) since both of them are located lateral of the axis of rotation (see Fig.1 and Fig.2). Of course this is all based on the standard kinesiological assumption that the body is in the so-called anatomical position and that all other hip movements besides rotation (e.g. hip flexion or adduction) are artificially prevented. (If you want, you may imagine someone hanging a slippery glass plate directly anterior and another one medial of the femur to prevent any movement of the femur in those directions, and then see how the femur adapts in response to a psoas contraction.)
Psoas pulls lesser trochanter forward
If lesser trochanter is pulled forward to rotate around the axis, then the greater trochanter will also be pulled forward
The biomechanical reasoning then goes like this
- The axis of femural rotation will be a straight line passing through the middle of the hip joint and the middle of the knee joint (see Fig.2). This line will pass slightly medial (!) of the lesser trochanter in the anatomical position. Which means that both the lesser trochanter and the greater trochanter are located lateral of the axis of femural rotation.
- If the pelvis is kept fixed and the femur mobile (yet only for rotational movement), a contraction of the psoas will tend to pull the lesser trochanter in the direction of the eminentia iliopubica where its fibers make a significant bend around the pelvic bone. In the so-called anatomical position this eminentia is located anterior in relation to the lesser trochanter (Fig.1). This means that a contraction of the psoas will tend to pull the lesser trochanter forward.
- Since both the lesser trochanter and the greater trochanter are located lateral in relation to the axis of rotation, a forward movement of the lesser trochanter will also move the greater trochanter forward in relation to the axis of rotation.
- A forward movement of the greater trochanter around the axis of rotation is an internal rotation of the femur.
Of course this rotational movement of the femur is a rather tiny and weak movement. The primary movement vectors of the psoas in terms of hip flexion/extension and adduction are definitely much stronger. So I don’t believe that it is very helpful to think of the psoas of being a possible direct mechanical cause for an internally rotated femur. Yet it seems quite clear that the common claim of many authors that a short psoas could be a direct mechanical cause for a chronic external femur rotation, is based on a faulty understanding. Also several popular osteopathic techniques now appear to be at least questionable in which the femur is rotated medially together with a hipjoint extension in order to "pre-stretch the psoas."
Psoas as a hip joint extensor
In her chapter on the psoas Ida Rolf drove many of us readers crazy by claiming that the contraction of the psoas could move the lumbar vertebrae backwards. This seems to be in contrast to the usual textbook description of the psoas as a hip joint flexor muscle (which would usually tend to tilt the pelvis anteriorly and thereby move the lumbars forward into a lordosis).
Interestingly, recent EMG measurements of the psoas have shown that in the human body a psoas contraction is often used to hold the lumbars posteriorly (see Calais-Germain, B., Anatomy for Movement, p.62; or Tortora and Anagnostakos, Principles of Anatomy and Physiology). How can that be?
Let's look at the body mechanics first in a position where it is easier to see, in the supine position with knees up, as for example for a ‘pelvic roll’ movement education. If the long fibers of the psoas then contract, they will tend to pull the lesser trochanter and T12 closer towards each other. Since the thorax is resting heavily on the floor and prevents the upper psoas attachment to move, it will be the lower attachment first that will move towards the other one. Under water or without gravity this would probably mean raising the whole femur off the ground in order to bring the lesser trochanter closer to T12. Yet in the gravity field the weight of the thigh and leg prevents quite a lot of resistance against this adaptation.
Fig.3: Client lying on floor with knees up. Contraction of the long psoas fibers can tilt the pelvis posteriorly.
Now lets suppose that the pelvis is relatively mobile for rolling movements around its transverse axis, i.e. to respond by rolling into a more anterior or posterior pelvic tilt position. Of course this assumes that the lumbar spine is not stiff and the hip joints are not held tight by spurious muscular holding patterns.
We then have the following mechanical conditions:
- the sacroiliac joint is posterior in relation to the acetabulum
- the femur is fixated (i.e. a moderate psoas contraction can not lift it off the ground)
- the lower thorax is fixated (a moderate psoas contraction will not lift T12 off the ground)
- the hip joints and lumbar joints are relatively mobile to allow the pelvis to roll on the floor into a posterior tilt direction
Simplified profile view of standing client. If femur and lumbo-dorsal junction are stabilized (against forward or backward movements of them), a contraction of the long psoas fibers may tilt the pelvis posterior, thereby bringing the two ends of the psoas closer towards each other.
In other words: If the thorax, feet and sacrum are kept resting on the floor, and the hip- and lumbar joints are kept relative mobile, then the pull of the lesser trochanter towards T12 will tend to roll the pelvis into a posterior direction so that the pubic symphysis and lesser trochanter move a bit closer towards the lumbo-dorsal junction. Which means this kind of psoas contraction will result in a hip joint extension.
The ‘Pelvic Roll’
In order to activate the psoas, I often use the following movement exercise. In supine position with knees up, the client is asked to slowly tilt her pelvis anteriorly and posteriorly.
1. Her usual first attempt to do so is by a visible contraction of the rectus abdominis during the posterior tilt direction.
2. When asked to put her soft hand on her belly and feel if she can find a smooth rolling movement which does NOT go along with abdominal contraction, most of the clients will then be able to learn to roll the pelvis by pushing/extending their feet slightly against the ground.
3. Finally as a third step some clients are able to further discover that they can roll the pelvis only "from within" without abdominal contraction and also without slightly increasing the pressure of their feet towards the ground. (According to the mechanical analysis of this article, this third style is seen as an antagonistically alternating action among psoas and iliacus: when the iliacus shortens, the pelvis rolls into a more anterior tilt position; and when the long fibers of the psoas shorten, the pelvis rolls into the opposite direction).
Yet quite likely those EMG measurements mentioned before were not done in lying on the floor, but rather in the upright position. Let's look at the biomechanics in this position. Again, a contraction of the long psoas fibers will tend to move the lesser trochanter and T12 closer towards each other. In the gravity field in standing the weight on top of the legs will prevent the femurs from moving easily. Also if the pelvis is shifted anteriorly in relation to the thorax (which is not uncommon in standing) the lumbo-dorsal junction area will have some mechanical resistance against being pulled forward due to the weight and resulting backwards vector of the thorax and upper body on top of it. Yet if the hip- and lumbar joints are free for a sagital tilting adaptation of the pelvis around its transverse axis, the pelvis might actually respond by a posterior tilt movement in order to bring the two attachment points of the long psoas fibers closer together.
Iliacus and psoas as antagonists? Yes, I think they often are. This also matches with the palpatory experience of several of us Rolfing instructors that in an anteriorly tilted pelvis (or innominate bone in a pelvic torsion) it is often the iliacus which is short and tight; whereas in a posteriorly tilted pelvis (or innominate) it is more often the long psoas fibers . Another level where this antagonism between psoas and iliacus makes sense is in relation to their enervation. The long psoas fibers are innervated from a relatively high level of the spinal cord via the ‘lumbar plexus’, together with the quadratus lumborum. Whereas the iliacus is innervated by a leg nerve, the femoral nerve, which it shares with the sartorius, tensor FL and rectus femoris. So in terms of their functional groupings via the nervous system, it seems that the iliacus is associated as a leg muscle which flexes the hip joint, whereas the long psoas is organized as a trunk muscle which stabilizes the lumbar spine.
Iliopsoas in relation to scoliosis and pelvic torsion
There have been several speculations claiming that an unilateral short psoas might be a frequent cause–or at least contributing factor--for a scoliosis. In terms of side-bending one would then suspect the lumbar spine to sidebend towards the side of the shorter psoas. In terms of rotation one would suspect the psoas fibers attaching at the lateral sides of the vertebrae to rotate this side more anteriorly, which would result in a general rotation of those vertebrae away from the side of the short psoas. Yet according to the generally accepted ‘Freyette’s First Law’ the lumbar vertebrae tend to rotate as a group in the opposite direction, i.e. with their vertebral bodies towards the side of their convexity (in relation to the observer). This rotation is also how just about all scioliotic spines appear in x-rays.
IF a short iliopsoas would indeed function as a significant factor for the sidebending of a lumbar scoliosis, this should be testable in the following way: with the pelvis kept immobile a hip joint flexion on the side of the short psoas would result in immediate decrease of the scoliosis, whereas bringing the femur back to line would increase the scoliosis again. Definitely not what one often sees in scoliotic people!
Further proof against the ‘iliopsoas theory’ of scoliosis comes from the fact that several surgical attempts to improve a severe scolosis by cutting the ‘short’ psoas have been reported to have never yielded to success.
Interesting detail: the right iliacus has usually a closer fascial connection with the intestines (specifically the ileo-cecal junction) whereas the right psoas has usually a closer fascial connection to the (descending) colon. This could mean that the usually common right-anterior pelvic torsion pattern could be in some cases an adaptation to visceral strain onto the musculo-sceletal system. Yet since any pelvic torsion will influence the "vertical distance" (i.e. difference in height) between the acetabulum and the sacroiliac joint, such a ‘viscerally triggered’ torsion pattern is only likely if in standing the sacral base appears higher on the side of the more anteriorly rotated ilium. Which in my experience is the case only in less than 30% of the pelvic torsion people.
Adductors in action
The latest edition of the book 'Muscles Alive - Their Functions Revealed by Electromyography' by J.V.Basmajian & C.J. De Luca provides some interesting insights about the adductors.
In the first edition of this book it was still written that "a surprising hiatus appears in our knowledge of the adductors. Forming an enormous mass on the medial side of the upper thigh, they must have considerable importance. In spite of this, their exact function is usually a matter of guess work."
Since then lots of EMG studies (surface and needle) have been done by Basmajian, V.Janda and other researchers. Here are some of their newer findings:
- Besides in adduction the adductors are active during MEDIAL hip joint rotation (except for the long vertical fibers of the add. magnus), thereby "settling a classic argument that usually leaned in the other direction."
- Additionally they are activated in other movements:
- the long portion of the adductor magnus acts like a hamstring muscle in hip extension
- the gracilis assists in knee flexion
- the pectineus often assists in hip flexion.
- Important difference between children and adults: In hip joint flexion almost all children engage the long adductors, whereas only few of the adults do so too.
- Same interesting difference with knee extension against resistance: most children use their adductors, and most adults don't. This feature I must admit has been a stimulating surprise for me. The authors suggest that "this labile response of the adductors is related to postural response" and "these muscles are facilitated through reflexes of the gait pattern rather than being called upon as prime movers.")
My preliminary conclusions for our Rolfing work:
1. The adductor muscles are not just functional adductor muscles of the leg (otherwise they would not be required to be built as large as they are). They are also designed to assist in hip extension (adductor magnus), knee flexion (gracilis) and hip flexion (pectineus). It does not make sense to try to mechanically "separate the adductors from the hamstrings" or from the quadriceps in terms of separating their myofascia. Anatomically the adductor magnus for example is supposed to have a firm gluing with the vastus medialis and a shared septum with the semimembranosus which I do not want to change.
2. In order to help a chronically internally rotated femur to rotate more out, it seems indeed useful to do some work on the adductors.
3. Also for chronically extended hip joint structures--i.e. posterior pelvis tilt--it makes sense to include some releasing work on the adductor magnus.
4. Some new inspiration for the 'knee forward-knee back' active client movement participation in our fourth hour side-lying Rolfing position while working on the adductors: Based on the above described difference between children and adults it makes sense to look for and to work for a 'knee forward'-movement without much adductor participation--which Ida Rolf would have probably called a "more mature" movement pattern. My work then focuses on a "functional differentiation" of the adductors during hip flexion. Yet for the hip extension movement such a functional differentiation is less clear or convincing.
5. I now also include KNEE extension movement in the fourth hour ("move your lower leg forward while keeping your upper leg where it is"). E.g. when doing this movement with my own "not so good leg" against resistance, I can detect the above described children's feature of an accompanying adductor contraction. Which could be a sign of a spurious or immature co-contraction. I now get some of my clients to do this knee extension movement (against slight resistance from one of my hands) while I work with my other hand on their adductor tissues and help these fibers to stay long and easy.
Tests for chronic shortness of iliopsoas and adductors
V. Janda has developed (or refined) the following tests for chronic myofascial shortness. In the so-called ‘Thomas-Test’ the client is lying with his trunk supine, sitbones just on the very edge of the table, both of his knees and hip joints maximally flexed with knees towards the chest (see Fig. 5). The client firmly holds both knees with his hands towards the chest (in order to stabilize the pelvis) and the examiner also leans with his waist firmly against the sole of the left foot of the client to support that pelvic stabilization. The examiner then takes the right knee of the client and slowly drops it in direction of the floor. If the knee drops back to about 180 degrees of hip joint extension this is considered ‘normal’. If the knee does not drop to that level, some hip joint flexor muscles with their related myofascia are too short.
Which one it is, can be determined by what happens at the end of that range of movement. Like a "Sherlock Holmes in the myofascial net" the practitioner can then make clear inferences from his observations.
- For example if the knee joint then extends (by extending the lower leg significantly forward), the rectus femoris is the "sinner."
- If the thigh abducts in this position, it might be the tensor FL .
- If neither adaptation happens, then the iliospsoas is short.
- If the femur rotates externally, then it is the sartorius.
The Thomas-Test position allows detailed analysis if a hip flexor shortness exists, and if so which of the following tissues are primarily short: iliopsoas, rectus femoris, tensor FL, or sartorius.
Working with an anterior pelvic tilt
If the client has a chronically anterior pelvic tilt in standing, this can have several mechanical reasons.
1. First of all it can just be a habitual postural collapse pattern which is NOT caused by any chronic myofascial shortness. This option will show up easily when the client lies supine on a flat firm surface and her hipjoints then extend all the way (i.e. to 180 degrees) with little or no lumbar lordosis.
2. Second option is that the anterior tilt is caused by a chronic shortness of the lumbar erectors which hold the lumbar spine in a lordotic pattern. This usually shows in the form of a lumbar lordosis in this same test position which does NOT decrease significantly when the knees of the clients are moderately elevated (by the examiner, not the client).
3. Third option is that the anterior tilt is caused by a hip flexor shortness. This will show by a decrease of the lordosis when the knees are passively elevated. Additionally it shows up in the ‘Thomas Test’ (see Fig. 5) in the inability of the femur to drop to a 180 degrees hip joint extension position. The same test then can be used to determine which one of the hip flexors is the primary restriction: iliopsoas, rectus femoris, tensor FL, or sartorius.
Depending on the proper diagnosis, different working strategies will be appropriate:
- In the first option the client needs postural and movement education (including sometimes tonifying tissue work).
- In the other options the identified shortened tissues need to be lengthened (by the Rolfer and/or the client).
Unfortunately there is no generally accepted test to distinguish between iliacus and psoas shortness. I developed one which helps me only sometimes: in the side-lying position with both hip- and knee joints flexed, I slowly lift the leg on the ceiling side of the client and move it backwards several times into hip extension. With my other hand I palpate with the thumb the middle of the sacrum plus simultaneously with the third fingertip the spinous process of L3 (or possibly higher). When at the end of the hip extension movement the pelvis tilts anteriorly, I try to palpate if the lumbars move exactly simultaneously with the sacrum or if they move slightly after the pelvic tilt.
To test the adductors the client lies supine with their right body side close to the edge of the table. The examiner abducts the client’s right leg with extended knee as far as easily possible (see Fig. 6). 40 degrees of hip joint adduction is considered ‘normal’. At the end of the range of motion in this direction one can then bend that knee joint 90 degrees (i.e. dropping the lower leg down while holding the knee) and see if this allows a significantly larger range of hip abduction. If this is the case, then it is the gracilis which is the most short. If not, one suspects the other (1-joint-) adductors.
Normal range of abduction is 40 degrees.
If the adductors are chronically short this might be less.
If the gracilis is the first restriction, then dropping that lower leg down (i.e. knee flexion) will yield a few more degrees of leg abduction than without.
"The greatest of all pleasures is the pleasure of learning"
Copyright 1998, R.Schleip
First published in ROLF LINES, November 1998
For a very valuable response to this article, please read M.Morrison: Further Thoughts on Femur Hip Rotation and the Hip Flexors Psoas and Iliacus
 Yet sometimes--especially in senior clients--the lumbar joints are too stiff to allow for a smooth pelvic roll movement.
 See, for example, Sultan, J., Towards a Structural Logic, Notes on S.I., 86/1, May 1986, p. 14-15.
 Whereas Freyette's Second Law (describing ipsilateral sidebending and rotation direction for single vertebral movements in the lumbar and thoracic spine) seems to be only accepted among osteopaths yet not within the much wider field of manual medicine in general. See Basmajian, J., Nyberg, R., Rational Manual Therapies, Williams & Wilkins, 1993, p. 295.
 Hugo A. Keim, H.A., Scoliosis, Clinical Symposia, Vol. 30, No. 1, 1978.
 See Schleip, R., Pelvic Torsion and Structural Alignment in the Gravitational Field, Rolf Lines, May 1996
 5th Edition, published by Williams and Wilkins. (Often referred to as "the bible of EMG research.")
 See Platzer, W., Color Atlas/Text of Human Anatomy, Vol. 1, Thieme, p. 236.
 Janda, Vladimir, Manuelle Muskelfunktionsdiagnostik, 1994, Ullstein-Mosby, Berlin; see also Tunnell, P.W., Muscle length assessment of tightness-prone muscles, Journal of Bodywork and Movement Therapies,
Vol. 2, No. 1, 1/1998.
. I.e. to about the same distance from the floor as the hip joint.
Yet, for example, most professional dancers or acrobats will not be satisfied with that range of motion.
 Additional signs for a tensor shortness are: slight knee joint extension (because most fibers of the iliotibial tract cross the knee
joint anterior to the knee axis in this position) and a clear visible increase of the tissue indentation of the fascia lata on the lateral thigh in this test position
 . Which will additionally abduct the leg. In my experience with this test the iliopsoas is the number one sinner, shortly followed by the rectus femoris. The tensor FL is the third most common one, and the sartorius the least frequent
 "Moderately" means lifting the knees only to a 20-40 degree decrease of the previous extended hip joint position. Otherwise—if lifting the knee much more (for example, towards the chest)--the pull of the gluteal fascia might tilt the pelvis and will distort this test.
 If the lumbar spine is very rigid anyway, this test will not work. Plus sometimes--e.g., if the leg is fairly heavy--the test is not that easy to perform with any palpatory certainty.
 See footnote 10