The sacroiliac (SI) joint: why does twisting hurt exactly at that dimple in the lower back?

The golden rule of anatomy: form follows function. Look at the structure of the SI and you can guess at once what it was made to do.

• The joint surfaces are NOT flat. The two surfaces are oriented obliquely, their surfaces rough with raised and hollow areas, interlocking like a three-dimensional jigsaw piece — 'hills and valleys' fitted into one another.
• That rough, close interlocking gives it very high mechanical stability, but in exchange it allows only an extremely small range of glide. You cannot have 'tight as a jigsaw' and 'mobile as a hip joint' at the same time.

What this gives you: use exactly this '3D jigsaw' image so students understand why forcing the joint to rotate or glide beyond its nature invites injury — this is the root of 'twisting hurts'.

This is where many teachers have it backwards. The texts classify the joint slightly differently (some call it cartilaginous, some a synovial 'gliding' joint, some 'a joint with very little movement') — but all converge on one point:

• The SI prioritises STABILITY. It allows only a minute amount of glide, of the order of a few millimetres, and that movement is INVOLUNTARY — you cannot 'deliberately rotate the SI joint' as you would a hip joint.
• The vertebrae of the sacrum are themselves fused and do not move relative to one another.

A figure to use as a hook in class: the joint glides only a few millimetres — so why does such a tiny movement cause severe pain? Precisely because the range is so small that, once it is forced past, the structure is instantly overloaded.

A correction of thinking for teachers: do not set the goal of 'opening the SI joint' as you would 'opening the hip'. For the SI, the right goal is STABILITY.

The normal physiological movement of this region has its own name, to be distinguished from 'pelvic tilt':

• Nutation: the base of the sacrum tipping slightly FORWARD.
• Counter-nutation: the base of the sacrum easing slightly BACK.

In walking the two sides can move in opposition: as the right leg steps forward, the right side of the pelvis tips slightly back and the left slightly forward. The range is a little greater in a long stride, a lunge, climbing stairs, or Hanumanasana — and a healthy SI joint permits that to happen physiologically.

A note on distinguishing where movement comes from: when we 'tilt the pelvis' (as in Cat–Cow) or the pelvis moves through space, the movement actually happens at the lumbar spine and the hip joints — NOT at the SI. The SI is all but immobile. Understanding this lets you teach students to take the large movement from the hips and the spine, rather than dumping it into the barely mobile SI joint.

This is the most directly practical explanation for a practitioner, and the place where a classic piece of bad advice is demolished:

• In a seated twist, if you PIN both sitting bones hard to the floor while the trunk still tries to rotate, then the whole torque is forced to 'eat' into the semi-immobile SI joint — putting a great deal of pressure and shear on the joint → which can produce aching pain.
• The shear mechanism: as the trunk turns, one side of the sacrum is pulled forward, opening the joint space on that side and compressing the other → producing a sliding force. Twisting to the end of the range using the arm as a lever brings that force to its maximum.
• The right way: let the two sitting bones glide GENTLY on the floor with the action (pelvis and sacrum rotating together), so the torque is distributed rather than poured entirely into the joint.

What this gives you: drop the cue 'pin the pelvis down and twist'. Replace it with: hold the pelvis actively steady with muscle while allowing it to travel with the turn — a subtle but important difference.

Recall: the stability of the SI relies on ligament. So:

• Repeated over-stretching (deep twists, maximal forward folds, very wide leg splits) can leave the sacroiliac ligaments lengthened and not readily springing back to their original length → the joint becomes less stable, with inflammation and pain.
• Histologically: ligament and tendon have very little elasticity. If the load passes the 'yield point', the tissue leaves the elastic zone (where it still recovers) for the plastic zone (damaged, and slow to recover).
• Signs of going past the range (teach students to recognise them): a sensation of twinge, numbness or pain right at the joint (not the sensation of stretch in the middle of a muscle belly), and having to hold the breath or grimace to maintain the pose.

An important nuance, to be corrected: research shows that a single passive stretch lengthens a ligament only temporarily before it returns to resting length; the real culprit in chronic dysfunction is repeated shear force and a loss of motor control, not simply 'one stretch that lengthened the ligament'. The message for a teacher: BOTH excessive range AND a lack of stability are dangerous.

This point is easily misread if you only take in half of it. With the SI joint there are two opposite traps:

• Pinning or locking the two sitting bones while twisting → pressure piled on to the joint structure → aching.
• Conversely, letting the joint move TOO MUCH (going slack, gliding excessively) → ALSO leads to aching.

Anatomy's conclusion: 'find the middle way for your own body'. The SI needs just enough stability and just enough mobility — neither locked rigid nor loose and displaced.

The number-one bewildering symptom: a twinge at one point around the PSIS, on one side only. The middle-way principle explains it:

• It is usually an asymmetry between the two sides: one joint stuck and stiff (hypomobile), forcing the other to compensate and move too much (lax). The side that 'eats' all the movement is the side reporting pain.
• So do not jump to 'the muscle on that side is tight, stretch it further' — you may be working the very side that is already too loose, and making it worse.

What this gives you: when a student hurts on one side while twisting, prioritise stability and balancing the force between the two sides, reduce the range, and observe which side is compensating — rather than pushing deeper.

The stability of a ligament is NOT fixed — it shifts with the body's biochemistry:

• Relaxin (secreted in quantity during pregnancy; some research suggests small amounts around ovulation) loosens the ligaments and the fibrocartilage supporting the pelvis, in preparation for birth.
• Stress and fatigue also weaken ligaments; hormonal changes across the monthly cycle can alter the stability.

The practical consequence: in these phases the SI joint is less stable → one must be careful not to twist or stretch excessively deep. This is exactly why the same person in the same pose is fine one day and in pain another. What this gives you: ask pregnant and postpartum students how they are and adjust the range; do not assume 'if they managed it last week they can manage it this week'.

• People with lax, supple joints are in fact the group MOST prone to SI pain — counter-intuitively. The more flexible side will 'eat' all the movement and glide excessively. They need to draw in and stabilise (a block between the knees, actively engaging the muscles around the pelvis), and should not be showing off range.
• Too much sitting also irritates the SI region and weakens the supporting muscles.
• An imbalance of force between the two sides is a seed of strain: for example, over-tight hamstrings pulling the pelvis into a posterior tilt → which can strain the SI joint; or one side of the pelvis tighter than the other → pulling the sacrum out of line.

What this gives you: always work symmetrically and evenly on both sides; with a supple student, teach 'hold' rather than 'reach further'; and remind students with tight hamstrings that they need releasing evenly so the pelvis is not pulled out of line.

Tight muscle around a painful region is usually the body's intelligent protective response — trying to immobilise and stabilise a joint that is unstable. Stretching hard into it usually makes it worse. The answer for a painful SI is STABILITY, not stretch. The intuition inverts: the more you stretch to ease the pain, the more it hurts.

Ideally you should feel NOTHING inside the SI joint while twisting. Any twinge or pinch inside the joint is suspect. The sensation of 'getting in deep' that many people take for a deep practice is in fact an alarm signal. The healthy sensation in a twist is the oblique abdominals and the back muscles 'working' — in the muscle belly, not in the joint space.

• Twisting does NOT 'wring out toxins'. The liver handles toxins very efficiently; there is no evidence that twisting 'detoxifies'. The real benefit of a twist is stimulating healthy intestinal motility, and psychologically you may picture 'letting tension go'. The implication: there is no biological reason whatever to force a twist to the point of SI joint pain.
• A 'crack' or 'pop' does NOT mean 'the joint has gone back into place'. A benign joint sound (occasional, painless) is quite different from a grinding noise accompanied by lasting pain — a sign of dysfunction. Depending on cracking it to make a sound can mask an underlying instability.

A key anatomical feature, which explains why stabilising the SI is harder than we think:

• NO muscle runs directly across the sacroiliac joint — unlike almost every other mobile joint (where a muscle crossing the joint helps control and hold it). Which means there is no dedicated 'SI joint muscle'.
• Stability has to come INDIRECTLY from the muscles attaching around the pelvis: gluteus maximus and medius (attaching close to the sacrum and the ilium), piriformis (attaching to the front of the sacrum), quadratus lumborum, the spinal erectors, and transversus abdominis (TVA) supporting from the front and sides. Gluteus maximus contracting isometrically also directly supports the sacropelvic region in many poses.

What this gives you: which is why, in a twist, you must ACTIVELY 'hold the pelvis solid' with the muscles around it (a light draw-in of the lower belly, the glutes moderately engaged) — rather than waiting for some muscle to lock the joint for you.

The routine for protecting the SI and the discs, agreed across the yoga anatomy texts:

1. Lengthen along the spine's axis first (axial extension). In rotation the disc is naturally compressed — so first rise tall and 'stack' each vertebra on the one below, creating space.
2. Only then rotate, and rotate chiefly from the CHEST, spreading up vertebra by vertebra — keeping the sacrum centred between the two pelvic wings.
3. Twist only as far as the MUSCLE takes you. Use the arm to HOLD the length of the spine, not as a lever to wrench further round.
4. Keep the pelvis level and let the sitting bones glide gently with the action; do not bear down heavily on the supporting hand; breathe evenly and never hold the breath to force it.

The takeaway for class: one compact instruction — 'Rise tall first, then turn; turn from the chest, do not wring at the waist.'

In the body's system of levers, the SI joint at the base of the spine acts as the fulcrum. When a student reaches the arm far out, lengthens the trunk and pours twisting force around that axis without the supporting muscles being strong enough, the torque on the SI rises sharply. The principle: the longer the lever, the greater the load on the fulcrum.

What this gives you: with a beginner, or anyone weak through the core, shorten the lever (hands close to the body, elbow against the knee) and prioritise building the muscles around the pelvis before letting them reach far or twist deep.

A 60-second experiment so that both you and your students really absorb the mechanism:

• Sit in a gentle twist (Bharadvajasana, or a cross-legged seated twist). Round one: deliberately PIN both sitting bones hard to the floor and twist to the end of the range — notice the sensation piling up in the lower back, around the dimple.
• Round two: rise tall first, let the two sitting bones glide gently and the pelvis travel with the turn, and rotate from the chest only as far as the muscle takes you — feel the SI region go markedly 'light'.

Add the landmark exercise: find the two PSIS dimples, place the fingers on them and then twist, to feel the joint region working. What this gives you: letting students experience the difference themselves is a far more persuasive way to teach safety than any explanation.

• Anyone with SI joint dysfunction or pain should avoid extreme twists and be careful with wide-leg poses (Trikonasana).
• In asymmetrical poses (Warrior, Triangle), holding one side too long can bring on SI discomfort → so change sides more often.
• Deep twists (Half Lord of the Fishes, Revolved Side Angle) are unsuitable for anyone with a spinal disc injury; anyone with lower-back pathology must practise very carefully, entering slowly and without jerking.
• Note the double edge: some poses that 'open' the SI region (the Prasarita Padottanasana type) are useful when needed, but require caution, because the joint tolerates only a small range.

What this gives you: settle on one line for the class to remember — a twinge inside the joint means come back, not push on; stability is your friend, and range is not the goal.