Gluteus Medius Anatomy

The gluteus medius is the "deltoid of the hip": a fan-shaped muscle that covers gluteus minimus and is the primary hip abductor. It is critical to normal gait because it stabilises the pelvis in the coronal plane during the single-leg stance phase of walking, and weakness leads to a Trendelenburg gait. Its anterior fibres also internally rotate and flex the hip; its posterior fibres extend and externally rotate it.

Origin. The muscle arises from the outer surface of the ilium between the anterior and posterior gluteal lines, and also from the overlying gluteal aponeurosis, the fascia that covers it.

View larger

Insertion. The fibres converge into a strong tendon that inserts on the lateral facet and the superoposterior facet of the greater trochanter. On an axial MRI through the trochanter, gluteus minimus attaches to the anterior facet and the principal gluteus medius insertion occupies the lateral facet beneath the iliotibial band, with tensor fasciae latae anteriorly and gluteus maximus posteriorly. That orientation is the map for interpreting axial MRI and planning anchor placement, and the intact facets are the reference against which tendon discontinuity, retraction and abnormal bursal fluid are judged.

View larger

View larger

The bursae. Three peritrochanteric bursae occupy distinct planes. The large trochanteric (subgluteus maximus) bursa lies superficial to the abductor tendons, deep to gluteus maximus; the subgluteus medius bursa lies deep to the medius tendon near its insertion, and the subgluteus minimus bursa deep to the minimus tendon. Naming the involved plane is more precise than calling every lateral-hip fluid collection trochanteric bursitis.

View larger

Innervation. The superior gluteal nerve (L4, L5, S1), from the sacral plexus, leaves the pelvis through the greater sciatic foramen superior to piriformis and runs laterally between gluteus medius and gluteus minimus, entering the deep surface of the muscle. It supplies medius and minimus and terminates in tensor fasciae latae. The other named nerves of the region are the inferior gluteal nerve, to gluteus maximus, and the posterior cutaneous nerve of the thigh.

Relations. Gluteus maximus covers the muscle posteriorly and tensor fasciae latae anteriorly; deep to it lie gluteus minimus and the superior gluteal vessels and nerve. The posterior border is the landmark for the posterior approach to the hip, and the anterior border is separated from tensor fasciae latae by the Watson-Jones interval. Below the muscle lie piriformis and the short external rotators (the superior and inferior gemelli, obturator internus, obturator externus and quadratus femoris), with the sciatic nerve emerging beneath piriformis.

View larger

View larger

View larger

View larger

Landmarks. Three palpable landmarks:

• Iliac crest: the superior border of the muscle
• Greater trochanter: the insertion, palpable on the lateral aspect of the hip
• ASIS: the anterior superior iliac spine

Injection safety. Draw a line from the PSIS to the greater trochanter. The injection should be superior and lateral to that line, in the upper outer quadrant, which avoids the sciatic nerve (medial and inferior) and the superior gluteal neurovascular bundle (deep but central). A deep injection, or a retractor wrongly placed in the posterior approach, can cause a sciatic palsy; identifying these landmarks is crucial to safe practice and avoiding iatrogenic nerve injury.

A spectrum, not a bursitis. Greater trochanteric pain syndrome was formerly called trochanteric bursitis. It is now understood as a spectrum from tendinosis through partial tear to complete tear, the "rotator cuff of the hip", with a pathophysiology similar to shoulder cuff disease. On imaging the tendon, not the bursa, is the key structure to inspect, because GTPS commonly reflects tendinopathy rather than isolated bursitis. Understanding the mechanical overload is key to successful conservative management.

Risk factors. Female sex (a wider pelvis and an increased varus moment), leg-length discrepancy and, above all, ipsilateral knee osteoarthritis, low back pain and iliotibial band tenderness: the neighbouring pathology that alters lower-limb and lumbopelvic mechanics. Obesity is the classic answer here and it is not supported: in Segal's population study BMI dropped out once knee osteoarthritis and back pain were adjusted for (see Global Epidemiology below).

Tears. Degenerative tears are the most common and occur at the insertion, the enthesis; traumatic tears are rare and associated with fractures. Fatty atrophy, graded by Goutallier, predicts poor outcomes after repair, as it does in the shoulder. On MRI, assess both tendons, tear thickness, retraction and muscle quality rather than reporting bursal fluid alone.

View larger

View larger

MRI grading of abductor tears. The grade guides the decision between endoscopic repair, open repair and muscle transfer:

• Grade 1: peritrochanteric oedema (bursitis or tendinosis)
• Grade 2: partial-thickness tear
• Grade 3: full-thickness tear, undisplaced
• Grade 4: full-thickness tear with retraction of less than 2cm
• Grade 5: massive retraction of greater than 2cm with fatty atrophy

Tear types. A four-type scheme is also used:

• Type I: tendinosis
• Type II: partial tear
• Type III: full-thickness tear
• Type IV: retracted tear with fatty atrophy

Trendelenburg test. The patient stands on one leg, the affected one. Normally the contralateral pelvis rises, because the abductors of the stance leg pull the ipsilateral pelvis down to level. In a positive test the contralateral pelvis drops: the weakness is in the stance-leg gluteus medius, not the side that falls, which is what "sound side sinks" means. The test must be held for 30 seconds to detect subtle weakness, which shows as fatigue.

Gait. Trendelenburg gait is the uncompensated form: standing on the affected leg, the pelvis drops on the contralateral (unsupported) side with each step. Duchenne gait is compensated: the trunk leans towards the affected side during stance, shifting the centre of gravity and reducing abductor demand. Bilateral weakness produces a waddling gait, often seen in hip dysplasia or myopathy.

Resisted abduction. In lateral decubitus the patient abducts the leg against gravity and resistance. Pain indicates tendinopathy; weakness indicates a tear or a nerve palsy.

Radiographs. An AP pelvis is checked for calcification at the insertion (calcific tendonitis) and for avulsion fractures. The fleck sign, a bony fragment superior to the greater trochanter, suggests avulsion.

Ultrasound. Excellent for dynamic assessment and for guided injection, and it can visualise fluid in the bursa and tendon tears.

MRI. The gold standard, and its sequences answer two different operative questions: what is torn, and whether the muscle remains worth repairing. T2 fluid signal at the insertion indicates a tear or bursitis. MRI also shows muscle quality and muscle belly atrophy, which raises the question of nerve injury; fatty infiltration is best assessed on the T1 axial sequence. T1 is essential because tendon continuity alone does not establish reparability: severe fatty change predicts limited strength recovery and may shift reconstruction toward transfer or salvage.

View larger

View larger

Nerve studies. EMG differentiates an L5 radiculopathy from a superior gluteal nerve palsy. An L5 root lesion also affects tibialis anterior and the peroneals, whereas a superior gluteal nerve palsy is isolated to gluteus medius, gluteus minimus and tensor fasciae latae. Denervation potentials (fibrillations and sharp waves) confirm active axonal loss.

Lateral hip and abductor-region pain has a wide differential. The examiner expects a structured separation of intra-articular, peritrochanteric, neurogenic and referred causes.

Anterolateral (Hardinge). Splits the anterior third of gluteus medius and vastus lateralis and gives an excellent view of the acetabulum. The superior gluteal nerve is at risk if the split extends more than 5cm proximal to the tip of the greater trochanter: running it too far risks denervating the anterior tensor fasciae latae and the anterior gluteus medius fibres, and the result is a persistent Trendelenburg limp. At closure, repair the tendon meticulously and restore the abductor sleeve securely to prevent a limp.

Posterior (Moore). Splits gluteus maximus, which is safe, and retracts gluteus medius anteriorly; a retractor that is too vigorous causes a traction injury to the superior gluteal nerve. The posterior approach relies on the integrity of the anterior abductor sleeve for stability.

View larger

Repair. Open repair uses a lateral incision with anchor fixation into the lateral facet, a double-row equivalent: traction sutures first demonstrate excursion and permit mobilisation, the proximal anchors establish the medial row, and distal anchors then compress the tendon across the greater-trochanter footprint. Endoscopic repair works through the peritrochanteric space, with an anterior portal lateral to the ASIS and a posterior portal at the tip of the greater trochanter. The technique is bursectomy first, identify the tear, decorticate the footprint, then suture anchors; knotless anchors are preferred to reduce prominence over the greater trochanter. The key operative distinction is whether the tendon is detached or retains a superficial layer that can be preserved, in which case a partial tear is treated transtendinously. The repair must be tension-free enough to protect during early rehabilitation.

View larger

View larger

A persistent Trendelenburg gait after hip surgery should trigger assessment of tendon continuity, muscle denervation and fatty atrophy rather than being dismissed as routine weakness.

• Persistent limp: failure of the repair or nerve injury
• Nerve injury: superior gluteal (abductor lurch) or sciatic (foot drop)
• Recurrence: re-tear rates are significant, 10-20%
• Heterotopic ossification: especially with lateral approaches

View larger

• Phase 1 (0-6 weeks): protected weight bearing on crutches; no active abduction; avoid adduction (crossing the legs), which stretches the repair
• Phase 2 (6-12 weeks): wean the crutches; aqua therapy; isometric abduction
• Phase 3 (3 months onwards): strengthening; return to sport and full activity

Conservative management succeeds in 70-80% of GTPS. Repair gives good pain relief in 90%, but strength recovery is variable and takes longer (1 year). Goutallier 3 or 4 fatty atrophy significantly lowers the success rate.

Treatment by pathology. Surgical intervention is reserved for those who fail 6 months of dedicated rehabilitation.

Injections. Corticosteroids provide short-term relief but may weaken the tendon. PRP is controversial but gaining popularity.

Global Epidemiology

• GTPS affects an estimated 10-25% of patients referred with hip/lateral thigh pain, with a population prevalence around 15% in women aged 50-79 versus roughly 6% in men (Segal NA et al., Arch Phys Med Rehabil 2007).
• Demographics: female predominance (odds ratio 3.37 in Segal's cohort; wider pelvis, larger abductor moment arm), peri- and post-menopausal age, leg-length discrepancy, and - the strongest associations in that multivariate model - ipsilateral knee osteoarthritis (OR 3.47), low back pain (OR 2.79), contralateral knee osteoarthritis (OR 1.74) and iliotibial band tenderness (OR 1.72). The pattern points at altered lower-limb and lumbopelvic mechanics rather than at the trochanter itself, which is why examination of the spine, both knees and gait matters more than another injection into a tender point.
• Two negatives from the same study that are worth carrying, because both are commonly taught as positives. Obesity was NOT independently associated with GTPS once age, sex, ITB tenderness, knee osteoarthritis and back pain were adjusted for (OR 1.10, 95% CI 0.80 to 1.52 comparing BMI ≥30 with under 25) - the apparent link is largely explained by the knee and back pathology that travel with it. And hip internal rotation range did not differ between those with and without GTPS, which is diagnostically useful in the other direction: preserved internal rotation does not argue against GTPS, whereas restricted and painful internal rotation should move hip osteoarthritis up the differential, not down it.
• Abductor tears are increasingly recognised as a degenerative enthesopathy ("rotator cuff of the hip") found incidentally in up to 20% of older women on MRI and in a notable proportion of THA candidates.

Side-by-Side Guideline & Society Positions

• The LEAP randomised trial (Mellor R et al., Br J Sports Med 2018) underpins the global shift towards education plus exercise as first-line therapy, outperforming a single corticosteroid injection at one year.

Registry & Approach Notes

• Arthroplasty registries (NJR UK, AOANJRR Australia, AJRR US) track surgical approach but not abductor integrity directly; the posterior approach dominates worldwide, with the direct anterior and lateral (Hardinge) approaches as the main alternatives.
• Comparative MRI data show the transgluteal (Hardinge) approach produces more measurable abductor damage and a higher rate of post-operative limp than the intermuscular direct anterior approach (Bremer AK et al., JBJS Br 2011) - a key trade-off against the Hardinge's low dislocation rate.

High- vs Limited-Resource Practice Variation

• Well-resourced settings: MRI staging, ultrasound-guided injections, endoscopic and arthroscopic abductor repair, and PRP/shockwave for tendinopathy.
• Limited-resource settings: Diagnosis rests on clinical examination (Trendelenburg, resisted abduction, single-leg stance) and plain radiographs; management is predominantly supervised exercise and selective open repair, with MRI and arthroscopy reserved for complex or surgical cases.

Corticosteroid injections. They provide rapid short-term relief but are inferior to exercise at one year and may be catabolic to tendon. The role and number of injections remains debated, with a clear trend away from repeated steroid (LEAP trial, Mellor 2018).

PRP and orthobiologics. Platelet-rich plasma and other biologics are increasingly used for gluteal tendinopathy, but high-quality comparative evidence is limited and results are inconsistent.

Endoscopic versus open repair. Both achieve good pain relief; endoscopic techniques offer less soft-tissue dissection but have a steep learning curve. No large randomised comparison establishes superiority, and selection (tear size, retraction, atrophy) drives outcomes more than technique.

The "safe zone" myth, and the direction of the risk. The classic 5cm rule is a fixed absolute distance applied to patients of every size, and Ekşioğlu's cadaveric series found the nerve-to-trochanter distance correlates with body height, so the margin is smallest in the shortest patients, who are the ones a 5cm split is most likely to reach. That is the opposite of the intuition that a bigger patient means a deeper, riskier dissection. The same study found the nerve outside the "safe area" in 22% of posterior regions (it was within it in all anterior regions), and its own conclusion is blunt: the safe area is not always safe. Note also that the literature conflicts, since other cadaveric work has found the distance independent of height, so the defensible position is not "worry in short patients only" but do not treat any fixed number as protection: keep the split as distal as the exposure allows, and if you need more proximal access, extend it as an anterior flap off the trochanter rather than by splitting further into the muscle belly.

Surgical approach for THA. The abductor-sparing advantage of the direct anterior approach (less measurable abductor damage on MRI) is weighed against its learning curve and complication profile; no approach is universally superior.