Introduction

In young adults, the choice for surgery or exercise treatment of meniscus tears does not lead to different outcomes in knee outcomes. This was previously confirmed by the DREAM trial. Several exploratory analyses have been conducted to determine whether subgroups exist for whom the results might be different. In this light, the current study sheds light on another possible subgroup: meniscus tears from traumatic incidents or sustained nontraumatically over time. It is relevant to consider this since nontraumatic meniscus tears are more common in older adults, whereas traumatic tears most often occur in young sporting adults. The current study therefore examined the differences between meniscus tear treatment choices for traumatic and nontraumatic tears.

Methods

In this research review, we discuss another exploratory analysis of the DREAM trial. In the original trial, the authors analyzed whether early surgery was superior to exercise. In one of our earlier research reviews, we summarized the exploratory analysis where it was investigated whether having mechanical symptoms or not affected the outcomes of the original DREAM trial and in another one, the authors wanted to know if the choice of treatment affected the progression of OA.

Today, we review another secondary analysis of the DREAM trial, which compared exercise to arthroscopic meniscal surgery in young individuals who had suffered a meniscus tear. In this study, the authors sought whether the meniscus tear treatment choices surgery or exercise affected knee outcomes when the analyses compared the etiology of the meniscus injury. Therefore traumatic meniscus tears were compared to nontraumatic tears.

Participants were eligible if they had an MRI-confirmed meniscus tear and were aged 18 to 40. Congenital discoid meniscus tears or misplaced bucket handle tears were not considered. 

They were randomly randomized to receive either exercise or arthroscopy (either a partial meniscectomy or a meniscus repair). Those randomized to receive exercise therapy could opt for later surgery if needed. The exercise consisted of a 12-week regimen with two supervised sessions each week. These participants were educated at the start and end of the program. In an earlier research review, we described the workout program’s components. However, in brief, the program consisted of:

• Warm-up
• Reducing swelling and increasing range of motion
• Neuromuscular exercises: sensorimotor control and functional joint stability
  • Knee bends, pelvic lifts, planks, side planks, stair climbing, outer thigh and inner thigh exercises using an exercise band, slide-exercise sideways lunges
• Strengthening exercises: increasing muscle strength and mass
  • ​​One-legged leg press, one-legged knee extension, one-legged knee flexion and kettlebell swings

The neuromuscular exercises were adjusted to each patient’s particular demands, with two to six levels of difficulty and two to three sets of 10-15 repetitions. Neuromuscular exercises were begun at a level established by the physical therapist based on visual assessment of the quality of the movement and sensorimotor control; whereas minimal exertion, discomfort during the exercise, and a sense of control over the movement were determined by the patient.

The strengthening exercises began with two sets of 15 repetitions, followed by three sets of 12 repetitions, three sets of 10 repetitions, and finally three sets of 8 repetitions. The strengthening exercises were developed using the +2 principle, meaning fewer repetitions were performed per set and more weight was added when the patient completed two extra repetitions in the final set.

In this particular analysis, the authors wanted to know whether people with a traumatic or nontraumatic meniscus tear form a subgroup that responds in a different manner than the people analysed in the original study. Traumatic tears were defined as tears sustained during a specific (such as kneeling, sliding and/or twisting of the knee or alike) or during a violent incident (like during sports, a crash, collision or similar). Nontraumatic meniscus tears were described as having developed slowly over time

The primary outcome was, like in the original trial, the between-group difference in the Knee injury and Osteoarthritis Outcome Score (KOOS), derived from the subscales pain, symptoms, function in sport and recreation and quality of life, but excluding the KOOS activities of daily living subscale. The between-group difference was measured at baseline, 3, 6 and 12 months, the latter being the primary endpoint.

Results

Sixty participants were randomized to surgery and 61 to exercise and education, bringing the total to 121 included participants. In the exercise therapy group, 42 patients sustained a traumatic tear, whereas 47 participants of the surgery had traumatic tears. Sixteen participants from the exercise group crossed over to surgery, but the number of traumatic and nontraumatic tears was equal. Seven patients of the surgery group finally did not receive surgery.

The KOOS did not show a between-group difference at 12 months when the participants with traumatic meniscus tears were compared to those with nontraumatic tears. These results were supported by the secondary outcomes. Both the participants in the surgery or exercise groups had clinically relevant improvements. These results indicate that the meniscus tear treatment choices do not have to be changed with different types of tear etiology. 

Questions and thoughts

A sensitivity analysis was conducted to exclude the traumatic meniscus tears sustained during a specific incident from those developed gradually. This was done since there is no consensus about the definition of a traumatic meniscus tear and the tears sustained during minor trauma may respond differently than those sustained during a violent trauma. While this is valuable to test the robustness of the results of the primary analysis, this in turn decreases the number of participants analyzed. This may limit the conclusions. 

Next to the intention-to-treat analysis, a per-protocol analysis was conducted to know whether differences in results emerged when considering the participants who had crossed over to surgery or did not participate in enough exercise sessions. In the per-protocol analysis patients randomized to exercise therapy were excluded if they participated in 17 or fewer of 24 exercise sessions (n=15) or crossed over to surgery (n=16) and the patients in the surgery group were excluded if not having surgery (n=8). This per-protocol analysis revealed no differences from the intention-to-treat analysis. 

Talk nerdy to me

The activities of daily living subscale of the KOOS questionnaire was not used in this trial, since it was reported to be not sensitive in a young study population. The minimally clinically important difference (MCID) was set at 10 points per subscale for defining a relevant improvement on the KOOS questionnaire. Importantly, this study only defined relevant differences when the 95% confidence interval did not include values below the MCID. Although this is a good approach, it may not be forgotten that this remains an exploratory analysis implying that the statistical power obtained from the original DREAM study was not valid to make conclusions for this study. As such, the results can give directions to future research but should be further confirmed first.

Therefore, the two WOMET subscales and the KOOS sports and recreational activities subscales seemed to favor surgery for traumatic meniscus tears, however, the confidence intervals were wide and non-significant, may have another result in a fully powered RCT. When the power is obtained to test this specifically, the conclusions of this exploratory analysis might change. Therefore, the results from this study can give direction, but can not be used as hard evidence until these are confirmed in the future. On the other hand, the results from this exploratory analysis confirmed the STARR trial results where arthroscopic partial meniscectomy was compared to exercise therapy in adults from 18-45 years. 

The results indicate that discussing whether or not to opt for surgery can be useful. Yet, in clinical practice, I find that many people immediately prefer surgery. The case for exercise therapy as the primary treatment choice may not yet be entirely conclusive, but the results of this study may help change the narrative.

Take home messages

Whether a meniscus tear originates from a traumatic event or develops gradually over time, the improvements from surgery or exercise therapy are similar. Thus, It is unnecessary to differentiate the treatment based on the meniscus tear etiology. This resulted from an exploratory analysis of the DREAM trial, which earlier confirmed that no difference in knee outcomes was seen in young individuals whether surgery or exercise was prescribed. This suggests that surgery and exercise therapy are reasonable treatment strategies for both types of meniscus tears, but also that symptom onset should not be the main driver of which treatment to opt for.