Knee after an injury: what the trials show about surgery and rehabilitation for meniscus and ACL tears

Most knee injuries are followed by the same two questions: is something torn, and does it need an operation? For the meniscus and the anterior cruciate ligament (ACL), those questions have now been tested in randomised trials, including a trial against sham surgery. The results are more specific than either "always operate" or "never operate", and they show which parts of standard practice rest on evidence and which rest on habit.

First question: is it a fracture?

Before the meniscus or ligaments come into it, the first decision after an acute knee injury is whether an X-ray is needed. The Ottawa knee rule was built for this: a short set of clinical findings (age 55 or older, tenderness at the head of the fibula, isolated tenderness of the kneecap, inability to bend the knee to 90 degrees, inability to take four steps both immediately after the injury and in the emergency department) that tells the clinician when an X-ray is worth taking. It was validated prospectively in adult emergency departments (Stiell et al., 1996).

A systematic review pooled six studies with 4,249 adults with acute knee injuries: the rule had a sensitivity of about 98.5% and a specificity of about 49% (Bachmann et al., 2004). In practical terms, a negative rule makes a fracture very unlikely, while a positive rule does not mean there is one — it means an X-ray is reasonable.

What an MRI finding does and does not mean

A torn meniscus on MRI sounds like a diagnosis that explains the pain. In middle-aged and older people, it often is not. In a community sample from Framingham, Massachusetts, 991 people aged 50 to 90 who were not selected for knee problems had an MRI of the right knee. Meniscal tears or meniscal destruction were found in 19% of women aged 50–59 and in 56% of men aged 70–90, and 61% of people with a tear had had no knee pain, aching or stiffness in the previous month (Englund et al., 2008).

This is an observational finding, and it does not say that tears never cause symptoms. It says that in this age group a tear on the scan is common enough without symptoms that the scan alone cannot settle whether the tear is the source of the pain.

Degenerative meniscal tears: the trials against sham surgery

Arthroscopic partial meniscectomy — trimming the torn part of the meniscus — was for years one of the most common orthopaedic operations. Three randomised trials tested it in middle-aged adults:

• Against sham surgery. In the Finnish FIDELITY trial, 146 patients aged 35 to 65 with symptoms of a degenerative medial meniscal tear and no knee osteoarthritis were randomised to partial meniscectomy or a sham procedure, with both patients and outcome assessors blinded. At 12 months, both groups had improved, and there was no meaningful difference between them in any primary outcome (Sihvonen et al., 2013).

• Against physical therapy, with osteoarthritis. In the US METEOR trial, 351 patients aged 45 or older with a meniscal tear and mild-to-moderate osteoarthritis were randomised to surgery plus physical therapy or to physical therapy alone. At 6 months, the improvement in function did not differ significantly; 30% of those assigned to physical therapy had crossed over to surgery within 6 months (Katz et al., 2013).

• Against physical therapy, without locking. In the Dutch ESCAPE trial, 321 patients aged 45 to 70 with a non-obstructive tear (no locking of the knee) were randomised to surgery or to 16 sessions of exercise therapy over 8 weeks. Over 24 months, exercise therapy was non-inferior to surgery for patient-reported knee function; 29% of the exercise group had surgery during follow-up (van de Graaf et al., 2018).

A meta-analysis of nine trials in middle-aged and older patients with degenerative knee disease found a small benefit for pain at 3 and 6 months that was not present at 1 to 2 years, no significant benefit for function, and harms that included deep vein thrombosis (Thorlund et al., 2015). A BMJ Rapid Recommendation panel issued a strong recommendation against arthroscopic surgery for nearly all patients with degenerative knee disease (Siemieniuk et al., 2017).

What these trials do not cover: they excluded or did not focus on knees that truly lock, and they studied degenerative tears in middle-aged adults, not acute tears in young people.

Meniscal tears in young adults

The Danish DREAM trial asked the same question in younger people. It randomised 121 adults aged 18 to 40 with MRI-verified meniscal tears to early surgery (partial meniscectomy or repair) or to 12 weeks of supervised exercise and education, with the option of surgery later. At 12 months, the difference between groups was not statistically significant; both groups improved, and about one in four participants in the exercise group went on to have surgery (Skou et al., 2022).

This is one trial of moderate size, and it is the first of its kind in this age group. It does not show that early surgery is superior to a strategy of exercise and education with surgery later if needed; it also does not show which individual tears would do better with surgery first.

ACL rupture: reconstruct early, or rehabilitate first?

Three randomised trials compared early reconstruction with rehabilitation first, with reconstruction later if needed:

• KANON (Sweden). 121 young, active adults with an acute ACL tear. At 2 years, the improvement in knee scores was almost identical between early reconstruction and rehabilitation with optional later reconstruction (Frobell et al., 2010). At 5 years, there was still no difference in knee scores or in radiographic osteoarthritis, and 51% of the rehabilitation-first group had had a delayed reconstruction (Frobell et al., 2013).

• COMPARE (the Netherlands). 167 patients aged 18 to 65 with an acute ACL rupture. At 24 months, the early-reconstruction group had a statistically better knee score, but the difference (5.3 points on a 100-point scale) was below what the authors considered clinically relevant; 50% of the rehabilitation-first group had a reconstruction during follow-up (Reijman et al., 2021).

• ACL SNNAP (UK). A different population: 316 patients with a non-acute ACL injury and persistent symptoms of instability. Here, surgical reconstruction as a strategy gave better knee scores at 18 months than rehabilitation, and 41% of the rehabilitation group had surgery within 18 months (Beard et al., 2022).

Read together, the trials point to a distinction rather than a single answer. After an acute ACL tear, starting with structured rehabilitation and reconstructing only if the knee keeps giving way led to similar patient-reported results in two trials, and about half of those participants had not had the operation by the end of follow-up. In a different population — non-acute ACL injury with persistent instability, not a group that had already completed rehabilitation — the UK trial favoured reconstruction as a strategy over rehabilitation first.

What is evidence and what is tradition

• Arthroscopic trimming of a degenerative meniscal tear in middle age. Not better than sham surgery or exercise therapy in the populations tested.

• Structured exercise therapy for a meniscal tear. Comparable results to surgery in middle-aged adults, and no significant difference in one trial in young adults.

• Early ACL reconstruction for every acute tear. Not superior on patient-reported scores to rehabilitation first in two trials of acute tears; about half of the rehabilitation-first participants did not have surgery during follow-up.

• Reconstruction for non-acute ACL injury with persistent instability. Favoured over a rehabilitation-first strategy in one large UK trial (patients were not required to have already completed rehabilitation).

• An MRI tear as proof of the cause of pain after 50. Tears are common in people without symptoms in this age group.

What these trials do not answer

• None of them applies to a knee that locks mechanically, to fractures, or to injuries of several ligaments at once; those were outside the trial populations.

• The outcomes are group averages of patient-reported scores. Some individuals in every trial did better with surgery, and the trials do not identify them in advance.

• Crossover was common in every rehabilitation-first arm. "Rehabilitation first" in these trials means a supervised programme with surgery available, not rest and waiting.

What to do with this

After a knee injury, the questions worth raising with a doctor are concrete: does the knee meet the criteria for an X-ray; does it lock or give way; what would a structured rehabilitation programme look like, and at what point would the decision about surgery be revisited? An MRI finding is one input into that decision, not the decision itself.

Sources

• Stiell IG et al. Prospective validation of a decision rule for the use of radiography in acute knee injuries. JAMA. 1996. PMID 8594242.

• Bachmann LM et al. The accuracy of the Ottawa knee rule to rule out knee fractures: a systematic review. Ann Intern Med. 2004. PMID 14734335.

• Englund M et al. Incidental meniscal findings on knee MRI in middle-aged and elderly persons. N Engl J Med. 2008. PMID 18784100.

• Sihvonen R et al. Arthroscopic partial meniscectomy versus sham surgery for a degenerative meniscal tear. N Engl J Med. 2013. PMID 24369076.

• Katz JN et al. Surgery versus physical therapy for a meniscal tear and osteoarthritis. N Engl J Med. 2013. PMID 23506518.

• van de Graaf VA et al. Effect of Early Surgery vs Physical Therapy on Knee Function Among Patients With Nonobstructive Meniscal Tears: The ESCAPE Randomized Clinical Trial. JAMA. 2018. PMID 30285177.

• Thorlund JB et al. Arthroscopic surgery for degenerative knee: systematic review and meta-analysis of benefits and harms. BMJ. 2015. PMID 26080045.

• Siemieniuk RAC et al. Arthroscopic surgery for degenerative knee arthritis and meniscal tears: a clinical practice guideline. BMJ. 2017. PMID 28490431.

• Skou ST et al. Early Surgery or Exercise and Education for Meniscal Tears in Young Adults. NEJM Evid. 2022. PMID 38319181.

• Frobell RB et al. A randomized trial of treatment for acute anterior cruciate ligament tears. N Engl J Med. 2010. PMID 20660401.

• Frobell RB et al. Treatment for acute anterior cruciate ligament tear: five year outcome of randomised trial. BMJ. 2013. PMID 23349407.

• Reijman M et al. Early surgical reconstruction versus rehabilitation with elective delayed reconstruction for patients with anterior cruciate ligament rupture: COMPARE randomised controlled trial. BMJ. 2021. PMID 33687926.

• Beard DJ et al. Rehabilitation versus surgical reconstruction for non-acute anterior cruciate ligament injury (ACL SNNAP): a pragmatic randomised controlled trial. Lancet. 2022. PMID 35988569.

This article is for information only and does not replace a consultation with a doctor.