Abstract
Background:
The mechanisms of failure and risk factors for failed meniscal surgery in children and adolescents have not been well described.
Purpose:
To investigate the risk factors, mechanisms of failure, and subsequent operative management for revision meniscal surgery in a pediatric population, as well as to identify the local incidence of failed meniscal surgery.
Study Design:
Case-control study; Level of evidence, 3.
Methods:
All patients younger than 20 years who had arthroscopic management for meniscal injuries at a single institution between 2008 and 2012 underwent retrospective review. Demographic data and intraoperative findings at the time of the initial surgery were documented. All patients undergoing a second procedure on the same meniscus were further analyzed. Multivariate logistic regression with purposeful selection was performed to identify independent risk factors for revision meniscal surgery.
Results:
Arthroscopic knee surgery was performed on 293 patients and 324 menisci, including 129 primary repairs, 149 primary partial meniscectomies, and 46 discoid saucerizations ± stabilization. At a mean of 40 months (range, 19-62 months) after surgery, 13% of all menisci required a revision procedure. The primary repair cohort had the highest failure rate (18%), followed by the primary discoid saucerization cohort (15%) and the partial meniscectomy cohort (7%). Multivariate analysis indicated that meniscal repair was predictive of retear (odds ratio, 2.04 [95% CI, 1.01-4.1]; P = .046), and children with an open physis and a bucket-handle tear had the highest retear rate of 46% (P = .039). Independent variables shown to have no significant relationship to revision meniscal surgery included age, sex, body mass index, extremity side, laterality (medial-lateral), time to repair, tear location, and associated ligament reconstruction. The most common indication for revision surgery was an acute reinjury during intense physical activity. Revision procedures were performed at a mean of 14 months after the index procedure, and the majority of failures (83%) were identified within 1 year. Of patients undergoing a revision surgery, 44% underwent a further debridement, whereas 56% underwent a repair.
Conclusion:
The success rate of meniscal surgery is 87% in children and adolescents. The revision rate was higher when repair was attempted in the index procedure, particularly in those children with open physes and bucket-handle tears. Most failures are the result of an acute reinjury within 1 year, and nearly half will require debridement of the retorn meniscus.
Meniscal tears continue to be a leading indication for knee surgeries, representing about half of all knee arthroscopies. 15 In the pediatric and adolescent populations, multiple factors account for the rising incidence of meniscal injuries, including increasing participation in athletics, expanding use of magnetic resonance imaging (MRI), and earlier diagnosis by physicians.2,4,11 The importance of functional menisci cannot be minimized, as they play a critical role in shock absorption, load sharing, reduction of contact stresses, and stability within the knee joint.10,16,18,29 Indeed, early degenerative changes and osteoarthritis become clinically apparent with partial or complete meniscectomies.9,30,34
Few studies have examined the clinical outcomes of meniscal repairs, partial meniscectomies, and discoid saucerizations in younger patients. Although excellent short-term results have been reported, there have been little data specifically looking at failed meniscal surgeries in this population. 35 We previously reported on a cohort of 293 patients who underwent a total of 324 operations for torn menisci. 31 The purpose of the current study was to follow the same cohort to investigate the risk factors, mechanisms of failure, and subsequent operative management for revision meniscal surgery in a pediatric population, as well as to observe the incidence of surgical failure.
Methods
After approval by our institutional review board, a retrospective chart review was performed of all patients younger than 20 years who underwent arthroscopic knee surgery for meniscal injury at a single institution between October 2008 and July 2012 by 2 pediatric sports surgeons. 31 Patients with a prior meniscal procedure were excluded from the study. Demographic and injury data from the index procedure, including sex, age, skeletal maturity, tear chronicity, and body mass index (BMI), were documented. Operative reports and intraoperative photographs were used to assess the tear pattern and location, as well as all concomitant procedures and injuries. Tear patterns were classified into the following categories: discoid, vertical, bucket-handle, radial, oblique (parrot beak), horizontal, fray, root detachment, or complex. Patients who were treated within 3 months of injury were considered to have an acute injury, and those treated after 6 months were considered to have a chronic injury. To more accurately differentiate acute and chronic tears, those treated between 3 and 6 months from the injury were not included in either chronicity group because injury dates were often estimated by patients or family members; 93 patients fell into this subacute group.
All patients undergoing an additional procedure on the same meniscus through December 2013 were further analyzed. Revision patients were placed into subgroups based on their postoperative course, either nonacute surgical failures or acute reinjuries of the same meniscus. The nonacute cohort was defined as those patients who required an additional operation due to persistent symptoms after initial surgery with no interval period of symptom relief. The acute reinjury cohort was defined as those who either had a period of symptom relief followed by recurrence or those who sustained an identifiable traumatic reinjury (whether through athletic activity or activities of daily living). This second cohort then had the mechanism, date of reinjury, and activity level documented. Univariate logistic regression was performed to identify independent variables for revision meniscal surgery, followed by multivariate logistic regression with purposeful selection. 5
Results
Between October 2008 and July 2012, primary meniscal repair, primary partial meniscectomy, or primary discoid saucerization with or without stabilization was performed on a total of 324 menisci in 293 patients. At a mean of 40 months (range, 19-62 months) after surgery, a revision procedure was performed on 41 menisci (13%) in 38 patients (Table 1). The mean time to revision was 14 months (range, 4-58 months) after the index procedure. The primary repair cohort had the highest failure rate (18%), followed by the primary discoid saucerization cohort (15%) and the primary partial meniscectomy cohort (7%) (Table 2).
Patient Characteristics at Index Procedure a
Results are reported as n (%), unless otherwise indicated. Counts and percentages are based on total number of tears; patients with bilateral tears were counted twice. BMI, body mass index.
Revisions Based on Primary Intervention a
Results are reported as n (% of total for that intervention).
The leading mechanism of failure was acute reinjury (31 menisci; 76%), most of which occurred within the first year (77%). Twenty occurred after full clearance by the treating surgeon. Of these, 17 occurred during strenuous physical activity, and 3 occurred during activities of daily living. Meanwhile, 11 acute reinjuries were sustained during the postoperative rehabilitation period: 3 were noncompliant with activity restriction, 3 occurred while performing physical therapy exercises, and 5 occurred accidentally while standing, walking, or stumbling.
Persistent symptoms after the index procedure were reported in 10 cases and were considered nonacute surgical failures. Five of these failures had been treated via meniscal repair, while the other 5 had undergone partial meniscectomy.
Each of the 7 failed discoid menisci that underwent a saucerization was combined into either the repair or partial meniscectomy cohorts based on stabilization method for subsequent statistical analyses. Three discoid saucerizations had tears that were stabilized with sutures and combined into the repair cohort; the remaining 4 were combined into the partial meniscectomy cohort. In total, 18% of all repairs including discoid stabilizations went on to revision surgery compared with 9% of all partial meniscectomies (odds ratio, 2.04 [95% CI, 1.01-4.1]; P = .046).
Revision surgeries were performed at a mean 14 months after the index procedure, consisting of 23 (56%) repairs and 18 (44%) partial meniscectomies. Concomitant anterior cruciate ligament (ACL) reconstruction was performed in 10 (24%) of the revision operations. Eight of these were revision ACL reconstructions.
Isolated meniscal reinjury, determined by physical examination and/or MRI, was the primary indication for 27 of the 41 revision operations. Combined ACL and meniscal injury was suspected preoperatively in 7 cases, all occurring within the acute reinjury group, while 3 injured menisci were identified only during ACL reconstruction. Four other meniscal injuries were incidental findings of procedures for other knee pathologic conditions, including symptomatic hardware, ACL cyclops lesion, chondral lesion, and contralateral meniscal tear.
Independent variables shown to have no significant relationship to revision meniscal surgery included age, sex, BMI, laterality (of knee or medial/lateral meniscus), growth plate status, time to repair, and prior ligamentous injury. Table 3 summarizes the meniscal injuries encountered at the revision surgery compared with the primary surgery. In addition, 93% of the tears were observed in the same location of the meniscus as that of the original tear. While no single tear classification was identified as a significant risk factor, multivariate analysis revealed that skeletally immature children with bucket-handle tears have a significantly higher rate of retears (46%; P = .039) (Table 4).
Index Surgery Meniscal Tear Type Compared With Revision Surgery
Effect of Skeletal Immaturity and Tear Pattern a
Results are expressed as n (% of total for that tear type).
Statistically significant (P = .039).
Considering the primary repair cohort alone, concomitant ACL reconstruction was performed in 63 of the 129 primary meniscal repairs (49%). Nine of these 63 combined procedures (14%) had a meniscal repair that failed, while 14 of the 66 isolated meniscal repairs (21%) required revision (P < .36). In the primary partial meniscectomy cohort, there were 66 concomitant ACL reconstructions (44%); 1 patient had combined partial meniscectomy and posterior cruciate ligament reconstruction. Four of these 66 (6%) went on to have a subsequent meniscal procedure.
Discussion
In this follow-up study of meniscal tears in children and adolescents, it is clear that revision meniscal surgery is not an uncommon procedure in this age group. At a mean of 40 months after a primary meniscal operation, 13% of all menisci required a revision procedure. The only single independent variable significantly associated with revision surgery was prior meniscal repair compared with partial meniscectomy. The likelihood of revision surgery is significantly higher in the skeletally immature patient with bucket-handle tear patterns. The majority of failures were due to acute reinjury and were mostly identified within 1 year.
The goal of any meniscal treatment is to restore the native biomechanics of the knee to alleviate discomfort and minimize the progression of osteoarthritis. When possible, meniscal repair is strongly favored as it has been shown to reduce the long-term incidence of osteoarthritis over partial meniscectomy.26,32 In the adult population, clinical outcomes and risk factors for arthroscopic meniscal surgery have been extensively studied. 27 The literature varies regarding the success rate of meniscal repairs, ranging from 56.5% to 95%, depending on the tear pattern, location, length of follow-up, and associated pathologic conditions.8,19,21,22,24,28 Lyman et al 20 reported on a large cohort in which 8.9% of meniscal repairs resulted in subsequent meniscectomies at a median of 3 years. A systematic review by Paxton et al 26 revealed that isolated meniscal repairs had a higher reoperation rate than did partial meniscectomies (23% vs 4%) in both the short term (<4 years) and long term (>10 years). This correlation between meniscal repair and need for revision operations is consistent with the findings in our study population.
More recent studies have focused on the outcomes of revision meniscal repair in adults, using validated meniscus outcome scales. 3 Voloshin et al 36 reported on a series of 18 patients who underwent repeat meniscal repair with a 72% clinical success rate and a mean follow-up of 7.33 years. Imade et al 14 reported on a series of 16 adult patients who underwent revision meniscal repair with a clinical success rate of 67% after a mean follow-up period of 41 months. Both studies concluded that revision meniscal repair should be considered an effective operative intervention. However, degenerative changes were observed in both studies, suggesting that revision repair may not adequately preserve the biomechanics of healthy menisci.
Despite the abundance of studies on the adult population, there remains a paucity of data on failed meniscal surgery in children and adolescents. Mintzer et al 23 were among the first to address clinical outcomes of meniscal repair in the younger population. With all but one of the repairs performed on the peripheral third of the meniscus, 100% of the 26 patients demonstrated clinically healed menisci. Vanderhave et al 35 reported on a series of 45 repaired menisci with a clinical success rate of 96% at 2-year follow-up. Both of these studies included concomitant ACL reconstructions among their cohorts, which some have postulated improved meniscal healing rates due to various causes.1,7,12,13 In contrast, Krych et al 17 studied isolated meniscal repairs and found a much lower success rate, with 17 of 47 repairs (38%) requiring revision and a mean follow-up period of 17 months. They determined that complex tear patterns and rim widths greater than 3 mm were associated with increased failure rates. While those studies focused on meniscal repairs, we reported on all meniscal tears seen in the pediatric population, which included repairs, partial meniscectomies, and discoid saucerization treatments. Overall, the failure rates in our patients and in the primary meniscal repair cohort fell between those reported in the literature.
ACL injuries are commonly seen in tandem with damage to adjacent structures, such as the menisci, cartilage, ligaments, and subchondral bone. It has been estimated that between 55% and 65% of all ACL injuries are accompanied by meniscal injuries. 7 Many have suggested that concomitant ACL reconstruction improves meniscal repair rates, prompting a number of investigations into this possible relationship.6,33 Both Lyman et al 20 and Wasserstein et al 37 found a significantly lower short-term rate of meniscal reoperations in adult patients who underwent concomitant ACL reconstruction. However, a meta-analysis by Nepple et al 25 concluded that there was no difference in failure rates between patients with intact ACLs and patients with reconstructed ACLs at 5-year follow-up. When evaluating the primary meniscal repair cohort, there was no statistically significant difference between failure rates in isolated meniscal repairs and combined meniscal repairs with ACL reconstruction, although there was a slight trend toward decreased revision rates with concomitant ACL reconstruction. We suspect that a larger sample size and longer follow-up may reveal the actual relationship between the 2 interventions in this population. A post hoc power analysis revealed that we would need 462 patients per group (N = 924) to have 80% power at P < .05.
The leading indication for revision meniscal surgery in our series was acute reinjury due to strenuous physical activity after clearance by the treating surgeon. More than a quarter of all reinjuries occurred during the postoperative period, either incidentally during activities of daily living or due to noncompliance with activity restriction. As such, patients should be cautioned on the risk of retear during the postoperative period and strongly advised to follow the surgeon’s rehabilitation time line. In addition, we found that meniscal repair failures in this patient population frequently occurred early, with 12% occurring within 3 months of surgery, 42% within 6 months, and 80% within 12 months. Given the nature of our study design, we cannot extrapolate regarding recommendations for how long patients should be protected before returning to sport.
There are a few inherent limitations to this study. The main weakness is related to retrospective study design, which results in the possibility of underestimating the need for revision meniscal surgery in this age group. Determination of treatment failures was based on patient follow-ups with the treating surgeon (and institution) and did not include those who sought subsequent treatment elsewhere. However, it should be noted that the large majority of children who seek care within our institution are contractually obligated to stay within our system based on their insurance (local, large independent physician associations and government insurer). Of interest, this limitation suggests that the true incidence of requiring revision meniscal surgery in children and adolescents is at least 13% but possibly higher. In addition, as the institution does not treat patients older than 19 years, many patients were likely to leave the region for college or seek further treatment with physicians who were not pediatricians. Finally, the diverse socioeconomic and cultural makeup of the population treated by this institution is quite unique to the region and may not be generalizable to the populations seen in other areas.
Conclusion
To our knowledge, this is the first study to report on the rates of revision surgery for meniscal repairs, partial meniscectomies, and discoid saucerizations in children and adolescents. Families should be consulted that patients’ risk of failing the initial surgery is significantly reduced once they are 1 year from the index procedure. The revision rate was higher when repair was attempted in the index procedure, particularly in those children with open physes and bucket-handle tears. As this study did not follow clinical outcomes, however, further long-term prospective studies are indicated in this younger population.
Footnotes
The authors declared that they have no conflicts of interest in the authorship and publication of this contribution.