Bilateral osteochondritis dissecans of the knee in pediatric and adolescent patients

Key points

1. 01Of 80 patients presenting with pain in only one knee, 12 (15%) had OCD in the opposite, asymptomatic knee on imaging.
2. 02Among patients with a silent contralateral lesion, 67% ended up having surgery on that asymptomatic knee.
3. 03There was no meaningful difference in age, sex, BMI, skeletal maturity, lesion location, size, or stability stage between the unilateral and bilateral groups.
4. 04Larger and deeper lesions were linked to more unstable (higher Hefti stage) disease.
5. 05Findings support routine bilateral knee imaging as screening, even when the other knee has no symptoms.

How it was conducted

Design

Retrospective single-institution case series (Level of evidence IV), patients diagnosed 2003 to 2016

Participants

80 patients aged 18 or younger with confirmed knee OCD and strictly unilateral knee pain, with contralateral knee imaging within 1 year (80 of 250 screened met criteria)

Groups

Unilateral OCD (n=68) compared with bilateral OCD (n=12)

Lesion assessment

Location by Cahill and Berg classification; size (width, length, depth) on radiograph and MRI; severity by Hefti classification (grades 1 to 2 stable, 3 to 5 unstable)

Primary outcome

Prevalence and characteristics of bilateral OCD, plus rates of surgical intervention

Statistics

Fisher's exact test, t tests or Wilcoxon rank sum; ANOVA for Hefti stage versus lesion volume and depth

What they found

• 80 patients: 63 males (78.8%) and 17 females (21.2%), mean age 13.1 +/- 2.2 years (range 8 to 18), mean BMI 21.2 +/- 4.6, mean follow-up 2.6 +/- 1.9 years.
• 12 patients (15%) had bilateral OCD on contralateral imaging; 5 of these contralateral lesions (42%) were stable on MRI.
• Primary lesion location: 57 (71.2%) medial femoral condyle, 11 (13.8%) lateral femoral condyle, 12 (15.0%) patella/trochlea; on lateral radiographs (excluding patella/trochlea) 58 (85%) region B and 10 (15%) region C.
• 20 of 80 primary lesions (25%) were stable on MRI.
• No significant differences between unilateral and bilateral groups: age (13.0 vs 13.3, p=0.724), male sex (76.5% vs 91.7%, p=0.422), BMI (21.1 vs 21.8, p=0.656), open growth plates (69.1% vs 75.0%, p=0.946), lesion location (p=0.604), and Hefti stage (p=0.186).
• Surgery on the symptomatic side: 52/68 (76.5%) unilateral vs 7/12 (58.3%) bilateral (p=0.337); 9/12 (75%) bilateral patients had surgery on at least one knee; 8/12 (66.7%) bilateral patients had surgery on the contralateral asymptomatic knee.
• Lesion volume and lesion depth each showed a positive association with Hefti classification (ANOVA p < 0.005 for both).
• Mean primary lesion dimensions: length 20.3 +/- 7.0 mm, width 14.5 +/- 4.6 mm, depth 6.20 +/- 2.3 mm.

Limitations

• Retrospective chart review subject to incomplete or inconsistent records.
• Selection bias from requiring contralateral imaging within 1 year, which may not represent all OCD patients.
• Very small bilateral subgroup (n=12), limiting the power to detect differences between groups.
• Single institution and Level IV evidence, so findings may not generalize.

Why it matters

For patients

If your child has OCD in one knee, the other knee can have a hidden lesion even without pain, so imaging both knees can catch problems early.

For clinicians

Routine bilateral knee imaging at presentation is reasonable for pediatric and adolescent OCD, since about 15% have a contralateral lesion and most such lesions later require surgery.

For readers

A small retrospective series suggesting silent contralateral OCD is common enough to justify screening both knees, though the bilateral group was too small for firm comparisons.