Prevalence of gluteus medius pathology on magnetic resonance imaging in patients undergoing hip arthroscopy for femoroacetabular impingement: asymptomatic tears are rare, whereas tendinosis is common

Key points

1. 01Among hips having arthroscopy for femoroacetabular impingement, 30.6% had gluteus medius pathology on MRI.
2. 02Asymptomatic pathology was present in 16.0% of hips, almost all of it tendinosis (14.3%), with only 1.7% partial-thickness tears.
3. 03No hips with a full-thickness tear on MRI were asymptomatic.
4. 04Severity of pathology rose with age: asymptomatic partial-thickness tears occurred in older patients than tendinosis (45.3 vs 39.4 years).
5. 05Patients 40 or older had a 2.11 times higher risk of asymptomatic pathology than those younger than 40.

How it was conducted

Design

Cross-sectional study, Level of evidence 3

Setting

Single high-volume hip preservation referral center, hip arthroscopy database February 2008 to January 2019

Participants

2851 hips (2452 patients) undergoing hip arthroscopy for femoroacetabular impingement with gluteus medius pathology on MRI

Groups

Symptomatic vs asymptomatic, defined by trochanteric tenderness, abductor weakness, Trendelenburg sign, or Trendelenburg gait on exam

Imaging

1.5-T MRI or MR arthrography graded as normal, tendinosis, partial-thickness tear, or full-thickness tear

Primary outcome

Prevalence of asymptomatic gluteus medius pathology

What they found

• Gluteus medius pathology was found in 871 of 2851 hips (30.6%) on MRI.
• Symptomatic pathology occurred in 414 hips (14.5%): 305 (10.7%) tendinosis, 99 (3.5%) partial-thickness tears, 10 (0.4%) full-thickness tears.
• Asymptomatic pathology occurred in 457 hips (16.0%): 408 (14.3%) tendinosis and 49 (1.7%) partial-thickness tears, with no full-thickness tears.
• Asymptomatic patients were younger than symptomatic patients (40.08 vs 46.89 years; P < .001).
• Among asymptomatic hips, those with partial-thickness tears were older than those with tendinosis (45.34 vs 39.41 years; P = .001).
• Patients 40 or older had a relative risk of 2.11 (1.80-2.50; P < .001) for asymptomatic pathology and 3.78 (3.09-4.61; P < .001) for symptomatic pathology versus those younger than 40.
• Interobserver reliability was substantial: Cohen k of 0.69 between MRI reports and lead author, and 0.72 between two authors.

Limitations

• Only patients undergoing surgery for femoroacetabular impingement were included, so results may not generalize to the average population.
• As a tertiary hip-preservation referral center, most patients had MRI for hip pain, limiting generalizability.
• Reliance on MRI review across scans from multiple outside centers introduces potential attentional and measurement bias.
• Cross-sectional design cannot determine which asymptomatic patients will develop symptoms over time.

Why it matters

For patients

An MRI report of gluteus medius tendinosis does not necessarily mean it is the source of your hip pain, but a tear is more likely to be clinically relevant.

For clinicians

Treat gluteus medius tendinosis based on full clinical evaluation rather than MRI alone, and weigh partial or full-thickness tears as more likely to be symptomatic.

For readers

Imaging-detected hip abductor tendinosis is partly a normal age-related degenerative finding, whereas asymptomatic tears are uncommon.