Can marathon running improve knee damage of middle-aged adults? A prospective cohort study

Key points

1. 01Subchondral bone marrow oedema in the medial tibial and femoral condyles improved significantly after the marathon (tibia p=0.011, femur p=0.082)
2. 02Lateral patellar cartilage damage worsened significantly (p=0.0005), suggesting the patellofemoral joint is most at risk
3. 03Iliotibial band friction syndrome increased markedly (p<0.0001) and prepatellar bursitis also increased (p=0.016)
4. 04Pre-existing asymptomatic meniscal tears, including bucket-handle tears, did not progress and did not affect finishing time
5. 05All radiological changes were asymptomatic; KOOS scores for pain, symptoms, and function showed no significant change

How it was conducted

Design

Prospective longitudinal cohort study with a comparison non-runner group

Participants

82 healthy, sedentary, novice middle-aged adults (median age 44, range 25-73) registered for the 2017 London Marathon; 71 completed the marathon and both MRI scans, 11 non-marathon runners also rescanned

Intervention

4-month standardised gradual marathon training programme followed by running the London Marathon

Imaging

Bilateral 3T MRI of both knees 6 months before and approximately 2 weeks after the marathon, graded by two blinded senior musculoskeletal radiologists using validated scoring systems

Primary outcomes

Changes in MRI scores for menisci, cartilage, subchondral bone marrow, tendons, ligaments, and bursa; KOOS self-report questionnaire

What they found

• Medial tibial subchondral bone marrow oedema improved significantly: 10 lesions improved, p=0.011
• Femoral condyle subchondral bone marrow oedema showed a trend toward improvement: 9 lesions improved, p=0.082
• Lateral patellar cartilage worsened significantly: 12 new or worsened lesions, p=0.0005
• Iliotibial band friction syndrome increased significantly: 15 new lesions, p<0.0001
• Semimembranosus tendon injuries increased: 6 new insertional injuries, p=0.016
• Prepatellar bursitis increased: 7 new lesions, p=0.016
• No significant change in meniscal tear prevalence; only 1 of 142 knees showed a new meniscal tear after the marathon
• KOOS scores showed no significant change for symptoms (p=0.981), pain (p=0.121), daily activity (p=0.303), sport and recreation (p=0.133), or quality of life (p=0.096)
• 67% of marathon runners reduced BMI during training; median BMI fell from 25.2+/-3.6 to 24.9+/-3.5 (p=0.009)
• Prior to the marathon, 65% of marathon-runner knees already had cartilage damage, 41% had bone marrow oedema, 42% had tendon injuries, and 42% had ligamentous lesions, all asymptomatic

Limitations

• MRI reporting carries inherent subjectivity bias despite use of two independent radiologists and consensus scoring
• Pre-study lifestyle and activity data were not available, relying on self-reported sedentary status at recruitment
• The comparison non-marathon group was small (n=11) and exact dropout timing was unknown, limiting control group interpretations
• Follow-up was only about 2 weeks post-marathon, so it is unknown whether worsened lesions resolve or persist long-term

Why it matters

For patients

Middle-aged novice runners can be reassured that running a marathon does not cause global knee deterioration and may actually benefit the main weight-bearing parts of the knee, but should be aware of elevated risk of patellofemoral and iliotibial band problems during training.

For clinicians

Pre-existing asymptomatic meniscal tears did not progress and did not impair performance, supporting conservative management; injury prevention programmes targeting the patellofemoral joint should be prioritised during marathon training.

For readers

This is the largest 3T MRI study of novice marathon runners to date and provides the most robust evidence that marathon running produces mixed but mostly asymptomatic structural changes in the knee, with potential protective effects on subchondral bone in the main weight-bearing compartments.