- Email: [email protected]
Age of 40 Years or Older Does Not Affect Meniscal Repair Failure Risk at 5 Years
Age of 40 Years or Older Does Not Affect Meniscal Repair Failure Risk at 5 Years Sarah Poland, B.A., Joshua S. Everhart, M.D., M.P.H., Walter Kim, M.D., M.P.H., Kent Axcell, Robert A. Magnussen, M.D., M.P.H., and David C. Flanigan, M.D.
Purpose: To compare meniscal repair failure rates in patients aged 40 years or older versus patients younger than 40 years. Methods: A total of 276 patients underwent meniscal repair surgery by a single sports medicine fellowshipetrained surgeon between 2006 and 2012 and were eligible for study inclusion. Patients were followed up for meniscal repair failure, defined as meniscectomy, repeated meniscal repair, or total knee arthroplasty. Logistic regression analysis was used to determine the risk of failure while controlling for potential confounding variables including body mass index, sex, anterior cruciate ligament status, time from injury to surgery, number of implants used, tear pattern, and chondral status at the time of the repair. Results: Among the 276 eligible patients, 221 (80%) were successfully contacted for follow-up at an average of 5 years after surgery. Of these patients, 56 were aged 40 years or older (mean, 47.2 years; standard deviation [SD], 5.3 years) and 165 were younger than 40 years (mean, 24.7 years; SD, 6.7 years). The overall meniscal repair failure rate over a 5-year period was 20%. Among patients aged 40 years or older, the failure risk was 18% versus 21% in patients younger than 40 years. After adjustment for confounding variables, age of 40 years or older was not associated with increased failure risk (adjusted odds ratio, 0.83; 95% confidence interval, 0.36-1.81; P ¼ .65). The mean time to failure tended to be shorter in older patients, at 16.9 months (SD, 10.2 months) versus 28.5 months in the group younger than 40 years (SD, 23.3 months) (P ¼ .04). Conclusions: Age of 40 years or older is not associated with an increased risk of meniscal repair failure at 5 years, although a shorter time to failure was noted in this age cohort. Level of Evidence: Level III, retrospective comparative study.
D
amage to the meniscus of the knee is one of the most commonly encountered orthopaedic injuries.1 The menisci play a role in shock absorption and lubrication and enhance the ability of the articular cartilage to provide a smooth, lubricated surface in the joint.2 Meniscal tears were originally treated with surgical debridement and removal of the damaged portion. There is now wellestablished evidence that meniscectomy contributes to
From the College of Medicine (S.P.); Department of Orthopaedics, Wexner Medical Center (J.S.E., W.K., R.A.M., D.C.F.); Department of Biochemistry (K.A.); and OSU Sports Medicine (R.A.M., D.C.F.), The Ohio State University, Columbus, Ohio, U.S.A. The authors report the following potential conflicts of interest or sources of funding: D.C.F. receives support from Ceterix Orthopaedics, Conmed, DePuy Mitek, Histogenics, Smith & Nephew, Vericel, and Zimmer. Full ICMJE author disclosure forms are available for this article online, as supplementary material. Received April 10, 2018; accepted November 25, 2018. Address correspondence to David C. Flanigan, M.D., Jameson Crane Sports Medicine Institute, The Ohio State University Wexner Medical Center, 2835 Fred Taylor Dr, Columbus, OH 43202, U.S.A. E-mail: david.flanigan@ osumc.edu Ó 2018 by the Arthroscopy Association of North America 0749-8063/18448/$36.00 https://doi.org/10.1016/j.arthro.2018.11.061
increased articular cartilage damage and accelerated development of osteoarthritis later in life.3-7 Meniscectomy, however, continues to be a popular procedure, with 465,000 persons undergoing the procedure every year in the United States.8 With the development of various arthroscopic techniques to repair the meniscus, meniscal repair procedures doubled in frequency between 2001 and 2005, indicating a preference to salvage as much of the native meniscus as possible.9 Meniscal repair procedures have been associated with higher reoperation rates, as well as higher knee outcome scores, than meniscectomy.10 Multiple studies have reported on failure rates and patient-reported outcomes after meniscal repair.10-14 Repair failure typically results in subsequent meniscectomy, repeated meniscal repair, or total knee arthroplasty. Reported failure rates in the general population are variable and have ranged from 9%11 to 36%,13 with an average of 22% to 24%.14 Despite increasing research into meniscal repair, there is no conclusive evidence regarding outcomes of meniscal repair in the older population, specifically in patients older than 40 years. Less favorable outcomes have been reported in the older age population, with some studies
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol
-,
No
-
(Month), 2019: pp 1-6
1
2
S. POLAND ET AL.
reporting a higher incidence of failure in the age groups older than 25 years15 and older than 30 years,16 which has been attributed to decreased vascularity of the meniscus with increased age. Other studies have reported repair failure rates ranging from 9%17 and 13.5%18 to 27%.19 In studies directly comparing the patient population aged 40 years or older and that younger than 40 years, nearly equal failure rates have been reported between the 2 age groups, at 5.3% and 5.5%, respectively, as reported by Steadman et al.20 Lyman et al.17 reported a lower risk of meniscectomy after repair in the older patient population, possibly owing to lower overall activity levels and involvement in lower-demand physical activities. The purpose of this study was to compare meniscal repair failure rates in patients aged 40 years or older versus patients younger than 40 years. We hypothesized that both age groups would have equivalent failure rates at a mean of 5 years’ follow-up.
Methods Data Sources This study was approved by the Biomedical Institutional Research Board of The Ohio State University. A total of 276 patients underwent meniscal repair surgery by a single sports medicine fellowshipetrained surgeon (D.C.F.) between 2006 and 2012 and were eligible for study inclusion. An a priori power analysis determined that this sample population with an estimated 20% or lower rate of loss to follow-up is adequate to detect a 13% difference in failure rates between age groups with 80% power and a 95% confidence level. Only 1 surgical procedure was included per patient. If a patient underwent multiple meniscal repair operations, only the initial procedure was included. Patients undergoing meniscal repair with or without concurrent injuries or procedures were eligible. All root repairs were excluded from this analysis. Medical record data including all demographic, preoperative, intraoperative, and postoperative rehabilitation and complication data were included. Demographic data collected included age, height, weight, sports participation, past knee injuries, and surgical procedures. Operative time, presence of cartilage defects, ligament injury, and concomitant procedures were recorded as intraoperative data. Complications, reoperations, and failures were recorded for the postoperative period. Patient Selection, Surgical Technique, and Rehabilitation Isolated meniscal repairs were performed in patients with clinical symptoms of meniscal tears including joint-line pain and clicking or catching and the absence of significant concomitant degenerative joint disease (defined as joint space narrowing >50% on
weight-bearing flexion posteroanterior radiographs). At the time of arthroscopy, the decision for meniscal repair versus partial meniscectomy was based on surgeon assessment of tissue quality, tear vascularity, and surrounding cartilage status. In the setting of concomitant cruciate ligament reconstruction, meniscal repairs generally were performed on unstable tears and tears greater than 1.0 cm in length, although case-by-case discretion was used by the operating surgeon with consideration of tear pattern, tissue quality, and tear location including vascular zone and compartment (medial vs lateral). Small (<1.0 cm), stable peripheral vertical tears in the setting of anterior cruciate ligament (ACL) injury were frequently left in situ. Meniscal repair was performed with either an insideout or all-inside technique based on surgeon preference and tear location. Inside-out repairs were performed for posterior horn and body tears of the lateral or medial meniscus using standard posterolateral and posteromedial approaches, respectively, with double-loaded, straight, flexible needles with No. 2-0 braided suture (Smith & Nephew, Memphis, TN). All-inside repairs were performed with Fast-Fix implants or Fast-Fix 360 implants (Smith & Nephew). As an adjunct procedure, rasping of the meniscus and adjacent synovium was performed at the time of the repair to stimulate healing. After meniscal repair, all patients were kept noneweight bearing for 4 weeks with no knee flexion beyond 90 for 8 weeks. No resistive hamstring exercises were used for 8 weeks. In the setting of isolated meniscal repairs, jogging was allowed as early as week 10 and a return to sports was allowed as early as week 16. Aside from the initial weight-bearing and range-ofmotion restrictions to protect the meniscal repair, in patients who underwent meniscal repair and concomitant ACL reconstruction, rehabilitation was performed according to the Multicenter Orthopaedic Outcomes Network (MOON) protocol.21 Outcome Measures For the purpose of this study, meniscal repair failure was defined as a recurrent or unhealed tear on repeated magnetic resonance imaging (MRI), subsequent meniscectomy, no evidence of healing on repeated arthroscopy, revision meniscal repair, or total knee arthroplasty performed on the original surgically repaired meniscus. All patients who had recurrent or persistent symptoms of meniscal tear underwent a repeated knee MRI scan followed by diagnostic arthroscopy if the MRI findings were suggestive of an unhealed repair or a subsequent meniscal tear at a site other than the repair. In this series, there were no patients with persistent symptoms and abnormal repeated MRI scans who declined repeated arthroscopy. Manipulation under anesthesia, lysis of adhesions, and repeated arthroscopy with evidence of a healed meniscal repair were not considered failures. Meniscal
3
AGE AND RISK OF MENISCAL REPAIR FAILURE Table 1. Descriptive Statistics All Patients (N ¼ 221) Demographic characteristics Age, mean (SD), yr BMI, mean (SD) Sex, n Male Female Tear zone, n Vascular Avascular extension Avascular Tear pattern, n Longitudinal Radial Horizontal Complex Tear length, mean (SD), cm Time from injury to surgery, mean (SD), d Previous ipsilateral knee surgery, n Ligament status, n ACL injury PCL injury MCL or LCL injury Multiligament injury Grade 3-4 chondral defects present, n Concomitant procedure, n Repair technique, n All inside Inside out Meniscus repaired, n Medial Lateral Both No. of implants, mean (SD) Time to failure, mean (SD), mo
30.4 (11.7) 26.7 (9.5)
Patients Aged 40 yr (n ¼ 56) 47.2 (5.3) 28.8 (6.6)
Patients Aged <40 yr (n ¼ 165) 24.7 (6.7) 27.3 (5.9)
121 (54.8%) 100 (45.2%)
28 (50.0%) 28 (50.0%)
93 (56.4%) 72 (43.6%)
139 (62.9%) 81 (36.7%) 1 (0.5%)
28 (50.0%) 12 (21.4%) 0 (0.0%)
79 (47.9%) 42 (25.5%) 1 (0.6%)
P Value <.001 .18 .31
.44
.01 165 27 22 7 1.45 86 37
(74.7%) (12.2%) (10.0%) (3.2%) (0.51) (97) (16.7%)
38 6 11 1 1.23 94 11
(67.8%) (10.7%) (19.6%) (1.8%) (0.51) (102) (19.6%)
127 21 11 6 1.49 83 26
(77.0%) (12.7%) (6.7%) (3.6%) (0.51) (94) (15.8%)
163 1 7 2 29 212
(73.7%) (0.5%) (3.2%) (0.9%) (13.1%) (95.9%)
39 0 2 0 11 54
(69.6%) (0.0%) (3.4%) (0.0%) (19.6%) (96.4%)
124 1 5 2 18 158
(75.2%) (0.6%) (3.0%) (1.2%) (10.9%) (95.8%)
170 (76.9%) 51 (23.1%)
52 (92.9%) 4 (7.1%)
.06 .60 .73 .06
.11 >.99 .02
118 (71.5%) 47 (28.5%) .21
142 53 26 4.4 24.8
(64.3%) (24.0%) (11.8%) (3.3) (21.4)
38 15 3 3.5 16.9
(67.9%) (26.8%) (5.4%) (2.5) (10.2)
104 38 23 4.7 28.5
(63.0%) (23.0%) (14.0%) (3.4) (23.3)
.14 .04
ACL, anterior cruciate ligament; BMI, body mass index; LCL, lateral collateral ligament; MCL, medial collateral ligament; PCL, posterior cruciate ligament; SD, standard deviation.
surgery on the contralateral meniscus was not considered a failure. Cartilage defects that were determined to be Outerbridge grade 3 or 4 lesions at the time of arthroscopy were noted.22 Cartilage defects noted at the time of meniscal repair were treated with either debridement, chondroplasty, or microfracture. For the purpose of this study, meniscal tears were defined both by pattern (vertical or longitudinal, horizontal cleavage, radial, or complex pattern) and by vascular zone (vascular, vascular with avascular extension, or avascular). Information on tear length and number of sutures used was included as well. Root tears were excluded. Statistical Analysis All statistical analyses were performed using a standard software package (JMP Pro, version 12.2.0; SAS Institute, Cary, NC). Descriptive statistics were first generated for the entire data set. Simple logistic regression models were created for all potential predictors to assess for an unadjusted association with meniscal repair failure. A
multivariate logistic regression model was then created to assess the independent association between age of 40 years or older and meniscal repair failure status. Potential covariates included age, weight, sex, ACL status, number of implants used, tear pattern, tear vascularity, and chondral status at the time of the procedure. A backward selection was used with an exit criterion of a greater than .05. If there was evidence of confounding as defined as a change in the estimated effect size of greater than 15%23,24 for age of 40 years or older, the covariate was kept in the model regardless of the a level.
Results Descriptive Statistics Among the 276 eligible patients, 221 were successfully contacted for follow-up (80% follow-up [221 of 276 patients]).The sample population consisted of 100 men and 121 women at more than 5 years’ mean follow-up (mean, 64.1 months; standard deviation
4
S. POLAND ET AL.
Table 2. Univariate Risk Factors for Meniscal Repair Failure Individual Risk Factor Age 40 yr <40 yr Sex Female Male BMI Tear location Vascular zone Avascular extension Tear pattern Vertical Horizontal Radial Tear length Time from injury to surgery Previous ipsilateral knee surgery ACL injury Grade 3-4 chondral defects Repair technique All inside Inside out Side of repair Medial Lateral Both No. of implants Use of 5 implants (yes)
*
Unadjusted OR (95% CI)
P Value
Referent 1.24 (0.57-2.70)
NA .57
Referent 1.04 (0.54-2.01) Per 1.0: 1.03 (0.95,1.10)
NA .90 .48
Referent 0.87 (0.31-2.46)
NA .80
Referent 3.34 (0.44-25.2) 1.48 (0.41-5.28) Per 1.0 cm: 0.91 (0.20-4.14) Per 1.0 mo: 1.11 (0.95-1.30)
NA .24 .55 .91 .18
1.11 (0.46-2.60)
.84
0.94 (0.44-1.98) 0.61 (0.32-1.19)
.88 .15
Referent 0.66 (0.29-1.55)
NA .33
Referent 1.50 (0.64-3.53) 0.94 (0.20-4.46) Per 1.0 implant: 1.05 (0.95-1.15) 1.11 (0.53-2.30)
.35 .94 .38 .79
ACL, anterior cruciate ligament; BMI, body mass index; CI, confidence interval; NA, not applicable; OR, odds ratio. *Likelihood ratio c-square test.
[SD], 33.7 months). Of the 221 patients, 56 (25.3%) were aged 40 years or older (mean, 47.2 years; SD, 5.3 years) and 165 (74.7%) were younger than 40 years (mean, 24.7 years; SD, 6.7 years) at the time of surgery. Overall, 173 patients (78.2%) had a concurrent ligamentous injury along with their meniscal tear. A total of 163 (73.7%) had isolated ACL tears, 7 (3.2%) had isolated medial or lateral collateral ligament tears, 1 (0.5%) had an isolated posterior cruciate ligament tear, and 2 (0.9%) had multiligament injuries (Table 1). The mean meniscal tear length was 1.45 cm (SD, 0.51 cm), and the mean time from injury to surgery was 86 days (SD, 97 days). Of the meniscal tears, 62.9% (139 of 221) were entirely within the vascular zone, 36.7% (81 of 221) had an avascular extension component, and 0.5% (1 of 221) were avascular. A total of 76.9% of the meniscal tears (170 of 221) were repaired with the all-inside technique, whereas the other 23.1% (51 of 221) were repaired with the inside-out technique. Meniscal Repair Failures and Complications The overall meniscal repair failure rate was 20% (45 of 221 patients) (Table 1). Of the failures, 27 underwent
a subsequent meniscectomy of the repaired meniscus, 10 underwent a repeated meniscal repair, 6 underwent an ACL reconstruction along with a subsequent meniscectomy or meniscal repair, and 2 underwent an unspecified meniscal procedure. Two complications occurred, both in the group aged 40 years or older, consisting of 1 reported deep venous thrombosis and 1 emergency department visit for uncontrolled pain postoperatively. Among patients younger than 40 years, the failure rate was 21% (35 of 165) versus 18% (10 of 56) in the group aged 40 years or older. Of the 45 patients who were considered to have failures at follow-up, the mean time to failure was 24.8 months (SD, 21.4 months). Unadjusted Risk Factors for Meniscal Repair Failure In the univariate analysis, no variables were significantly associated with an increased risk of meniscal repair failure (Table 2). The mean time to failure was significantly shorter in patients aged 40 years or older (mean, 16.9 months; SD, 10.2 months) versus patients younger than 40 years (mean, 28.5 months; SD, 23.3 months) (P ¼ .04, t test with unequal variances), but the risk of failure at a mean of 5 years’ follow-up was equivalent (odds ratio [OR], 1.24; 95% confidence interval [CI], 0.57-2.70; P ¼ .57, likelihood ratio c-square test). The presence of cartilage defects at the time of arthroscopy (P ¼ .15), sex (P ¼ .90), tear pattern (P ¼ .80), tear length (P ¼ .91), time from injury to surgery (P ¼ .18), and prior knee surgery (P ¼ .84) were not predictive of meniscal repair failure. In addition, concurrent ACL injury (P ¼ .88), weight (P ¼ .73), and use of 5 or more implants during surgery (P ¼ .79) were not predictive of failure in the univariate analysis. Independent Relation Between Age and Risk of Failure After we controlled for confounding variables including weight, sex, ACL status, number of implants used, tear pattern, and chondral status at the time of the procedure, age of 40 years or older was not associated with an increased risk of meniscal repair failure (adjusted OR, 0.83; 95% CI, 0.36-1.81; P ¼ .65) (Table 3). Use of 5 or more sutures or meniscal implants showed a trend toward higher failure rates (adjusted OR, 1.79; 95% CI, 0.97-3.22; P ¼ .06) and was kept in the multivariate model because of evidence of moderate confounding of the effect of age group on failure rates (31% change in effect size when number of implants was removed from the model).
Discussion We found no significant difference in meniscal repair failure rates at 5-year follow-up between patients aged 40 years or older and patients younger than 40 years. Failures tended to occur earlier in the group aged
5
AGE AND RISK OF MENISCAL REPAIR FAILURE Table 3. Independent Risk Factors for Meniscal Repair Failure Independent Risk Factor Age 40 yr Use of 5 implants
Adjusted OR (95% CI)* 0.83 (0.36-1.81) 1.79 (0.97-3.22)
P Valuey .65 .06
CI, confidence interval; OR, odds ratio. *Adjustment performed as needed for sex, weight, tear pattern, repair technique, and concomitant injuries. y Likelihood ratio c-square test.
40 years or older than in the group younger than 40 years (mean, 16.9 months vs 28.5 months), although failure rates at final follow-up were similar. A trend toward higher failure rates was observed with use of 5 or more sutures or implants; no other independent risk factors were identified, including cartilage defects, prior knee surgery, weight, sex, tear pattern, and ACL status. The aforementioned failure rate is consistent with other studies, which have found that older patients have a similar, if not decreased, rate of meniscal repair failure.17-20 In fact, in a study by Lyman et al.17 in which age was treated as a continuous variable, older age was shown be a protective factor against meniscal repair failure, given that older patients tend to be less active and return to physically demanding activities at a lower and slower rate versus their younger counterparts. A decreased activity level among older patients may place fewer mechanical demands on the meniscal repair. In our study, meniscal repair failures in patients aged 40 years or older tended to occur earlier than failures in younger patients. Early failures could potentially be due to lack of healing at the repair site, although this cannot be proved definitively with the current study design. With increased age, the vascular supply of the meniscus is diminished.25 This hypovascular environment could potentially affect the initial healing ability of our meniscal repairs, contributing to a shorter time to failure in this older age cohort. In 2 studies that noted a higher meniscal repair failure risk with increased age, Eggli et al.16 similarly suggested that meniscal healing may be affected by decreased vascularity of the meniscus with increased age whereas Buseck and Noyes15 did not provide a possible explanation in their article. Limitations A limitation of this study is loss to follow-up of a portion of the original cohort. Although a minimum 80% follow-up was achieved, greater than 95% followup is considered ideal to reduce risk of bias. Contact attempts for each patient were made a minimum of 3 times by E-mail, phone, and mail until the patient was determined to be lost to follow-up. Functional and subjective outcome scores were not included in this
analysis. Although our study has one of the largest samples of patients aged 40 years or older in the published literature, it is still underpowered to detect small differences in failure risk between age groups. In addition, the results are based on a series of patients from 1 institution and may be less generalizable than multicenter study results. Most meniscal repairs in this study occurred in patients with concomitant ACL reconstructions; although ACL status was not associated with meniscal repair failure risk, it is possible that there are age-specific differences in isolated meniscal repair failure risk that were undetected in this study. Finally, asymptomatic patients did not routinely receive repeated knee MRI scans to assess meniscal repair healing; it is possible that some asymptomatic patients may have unhealed repairs that were not accounted for in this study’s definition of repair failure.
Conclusions Age of 40 years or older is not associated with an increased risk of meniscal repair failure at 5 years, although a shorter time to failure was noted in this age cohort.
References 1. Clayton RA, Court-Brown CM. The epidemiology of musculoskeletal tendinous and ligamentous injuries. Injury 2008;39:1338-1344. 2. Fithian DC, Kelly MA, Mow VC. Material properties and structure-function relationships in the menisci. Clin Orthop Relat Res 1990:19-31. 3. Roos H, Laurén M, Adalberth T, Roos EM, Jonsson K, Lohmander LS. Knee osteoarthritis after meniscectomy: Prevalence of radiographic changes after twenty-one years, compared with matched controls. Arthritis Rheum 1998;41:687-693. 4. Aagaard H, Verdonk R. Function of the normal meniscus and consequences of meniscal resection. Scand J Med Sci Sports 1999;9:134-140. 5. Lanzer WL, Komenda G. Changes in articular cartilage after meniscectomy. Clin Orthop Relat Res 1990:41-48. 6. Henry S, Mascarenhas R, Kowalchuk D, Forsythe B, Irrgang JJ, Harner CD. Medial meniscus tear morphology and chondral degeneration of the knee: Is there a relationship? Arthroscopy 2012;28:1124-1134.e2. 7. Fairbank TJ. Knee joint changes after meniscectomy. J Bone Joint Surg Br 1948;30:664-670. 8. Kim S, Bosque J, Meehan JP, Jamali A, Marder R. Increase in outpatient knee arthroscopy in the United States: A comparison of National Surveys of Ambulatory Surgery, 1996 and 2006. J Bone Joint Surg Am 2011;93:994-1000. 9. Abrams GD, Frank RM, Gupta AK, Harris JD, McCormick FM, Cole BJ. Trends in meniscus repair and meniscectomy in the United States, 2005-2011. Am J Sports Med 2013;41:2333-2339. 10. Paxton ES, Stock MV, Brophy RH. Meniscal repair versus partial meniscectomy: A systematic review comparing reoperation rates and clinical outcomes. Arthroscopy 2011;27:1275-1288.
6
S. POLAND ET AL.
11. Perdue PS, Hummer CD, Colosimo AJ, Heidt RS, Dormer SG. Meniscal repair: Outcomes and clinical follow-up. Arthroscopy 1996;12:694-698. 12. Pujol N, Tardy N, Boisrenoult P, Beaufils P. Long-term outcomes of all-inside meniscal repair. Knee Surg Sports Traumatol Arthrosc 2015;23:219-224. 13. Tuckman DV, Bravman JT, Lee SS, Rosen JE, Sherman OH. Outcomes of meniscal repair: Minimum of 2-year follow-up. Bull Hosp Jt Dis 2006;63:100-104. 14. Nepple JJ, Dunn WR, Wright RW. Meniscal repair outcomes at greater than five years: A systematic literature review and meta-analysis. J Bone Joint Surg Am 2012;94: 2222-2227. 15. Buseck MS, Noyes FR. Arthroscopic evaluation of meniscal repairs after anterior cruciate ligament reconstruction and immediate motion. Am J Sports Med 1991;19: 489-494. 16. Eggli S, Wegmüller H, Kosina J, Huckell C, Jakob RP. Long-term results of arthroscopic meniscal repair. An analysis of isolated tears. Am J Sports Med 1995;23: 715-720. 17. Lyman S, Hidaka C, Valdez AS, et al. Risk factors for meniscectomy after meniscal repair. Am J Sports Med 2013;41:2772-2778.
18. Barrett GR, Field MH, Treacy SH, Ruff CG. Clinical results of meniscus repair in patients 40 years and older. Arthroscopy 1998;14:824-829. 19. Noyes FR, Barber-Westin SD. Arthroscopic repair of meniscus tears extending into the avascular zone with or without anterior cruciate ligament reconstruction in patients 40 years of age and older. Arthroscopy 2000;16:822-829. 20. Steadman JR, Matheny LM, Singleton SB, et al. Meniscus suture repair: Minimum 10-year outcomes in patients younger than 40 years compared with patients 40 and older. Am J Sports Med 2015;43:2222-2227. 21. Wright RW, Haas AK, Anderson J, et al. Anterior cruciate ligament reconstruction rehabilitation: MOON guidelines. Sports Health 2015;7:239-243. 22. Outerbridge R. The etiology of chondromalacia patellae. J Bone Joint Surg Br 1961;43:752-757. 23. Maldonado G, Greenland S. Simulation study of confounder-selection strategies. Am J Epidemiol 1993;138: 923-936. 24. Mickey RM, Greenland S. The impact of confounder selection criteria on effect estimation. Am J Epidemiol 1989;129:125-137. 25. Gray JC. Neural and vascular anatomy of the menisci of the human knee. J Orthop Sports Phys Ther 1999;29:23-30.
Purpose: To compare meniscal repair failure rates in patients aged 40 years or older versus patients younger than 40 years. Methods: A total of 276 patients underwent meniscal repair surgery by a single sports medicine fellowshipetrained surgeon between 2006 and 2012 and were eligible for study inclusion. Patients were followed up for meniscal repair failure, defined as meniscectomy, repeated meniscal repair, or total knee arthroplasty. Logistic regression analysis was used to determine the risk of failure while controlling for potential confounding variables including body mass index, sex, anterior cruciate ligament status, time from injury to surgery, number of implants used, tear pattern, and chondral status at the time of the repair. Results: Among the 276 eligible patients, 221 (80%) were successfully contacted for follow-up at an average of 5 years after surgery. Of these patients, 56 were aged 40 years or older (mean, 47.2 years; standard deviation [SD], 5.3 years) and 165 were younger than 40 years (mean, 24.7 years; SD, 6.7 years). The overall meniscal repair failure rate over a 5-year period was 20%. Among patients aged 40 years or older, the failure risk was 18% versus 21% in patients younger than 40 years. After adjustment for confounding variables, age of 40 years or older was not associated with increased failure risk (adjusted odds ratio, 0.83; 95% confidence interval, 0.36-1.81; P ¼ .65). The mean time to failure tended to be shorter in older patients, at 16.9 months (SD, 10.2 months) versus 28.5 months in the group younger than 40 years (SD, 23.3 months) (P ¼ .04). Conclusions: Age of 40 years or older is not associated with an increased risk of meniscal repair failure at 5 years, although a shorter time to failure was noted in this age cohort. Level of Evidence: Level III, retrospective comparative study.
D
amage to the meniscus of the knee is one of the most commonly encountered orthopaedic injuries.1 The menisci play a role in shock absorption and lubrication and enhance the ability of the articular cartilage to provide a smooth, lubricated surface in the joint.2 Meniscal tears were originally treated with surgical debridement and removal of the damaged portion. There is now wellestablished evidence that meniscectomy contributes to
From the College of Medicine (S.P.); Department of Orthopaedics, Wexner Medical Center (J.S.E., W.K., R.A.M., D.C.F.); Department of Biochemistry (K.A.); and OSU Sports Medicine (R.A.M., D.C.F.), The Ohio State University, Columbus, Ohio, U.S.A. The authors report the following potential conflicts of interest or sources of funding: D.C.F. receives support from Ceterix Orthopaedics, Conmed, DePuy Mitek, Histogenics, Smith & Nephew, Vericel, and Zimmer. Full ICMJE author disclosure forms are available for this article online, as supplementary material. Received April 10, 2018; accepted November 25, 2018. Address correspondence to David C. Flanigan, M.D., Jameson Crane Sports Medicine Institute, The Ohio State University Wexner Medical Center, 2835 Fred Taylor Dr, Columbus, OH 43202, U.S.A. E-mail: david.flanigan@ osumc.edu Ó 2018 by the Arthroscopy Association of North America 0749-8063/18448/$36.00 https://doi.org/10.1016/j.arthro.2018.11.061
increased articular cartilage damage and accelerated development of osteoarthritis later in life.3-7 Meniscectomy, however, continues to be a popular procedure, with 465,000 persons undergoing the procedure every year in the United States.8 With the development of various arthroscopic techniques to repair the meniscus, meniscal repair procedures doubled in frequency between 2001 and 2005, indicating a preference to salvage as much of the native meniscus as possible.9 Meniscal repair procedures have been associated with higher reoperation rates, as well as higher knee outcome scores, than meniscectomy.10 Multiple studies have reported on failure rates and patient-reported outcomes after meniscal repair.10-14 Repair failure typically results in subsequent meniscectomy, repeated meniscal repair, or total knee arthroplasty. Reported failure rates in the general population are variable and have ranged from 9%11 to 36%,13 with an average of 22% to 24%.14 Despite increasing research into meniscal repair, there is no conclusive evidence regarding outcomes of meniscal repair in the older population, specifically in patients older than 40 years. Less favorable outcomes have been reported in the older age population, with some studies
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol
-,
No
-
(Month), 2019: pp 1-6
1
2
S. POLAND ET AL.
reporting a higher incidence of failure in the age groups older than 25 years15 and older than 30 years,16 which has been attributed to decreased vascularity of the meniscus with increased age. Other studies have reported repair failure rates ranging from 9%17 and 13.5%18 to 27%.19 In studies directly comparing the patient population aged 40 years or older and that younger than 40 years, nearly equal failure rates have been reported between the 2 age groups, at 5.3% and 5.5%, respectively, as reported by Steadman et al.20 Lyman et al.17 reported a lower risk of meniscectomy after repair in the older patient population, possibly owing to lower overall activity levels and involvement in lower-demand physical activities. The purpose of this study was to compare meniscal repair failure rates in patients aged 40 years or older versus patients younger than 40 years. We hypothesized that both age groups would have equivalent failure rates at a mean of 5 years’ follow-up.
Methods Data Sources This study was approved by the Biomedical Institutional Research Board of The Ohio State University. A total of 276 patients underwent meniscal repair surgery by a single sports medicine fellowshipetrained surgeon (D.C.F.) between 2006 and 2012 and were eligible for study inclusion. An a priori power analysis determined that this sample population with an estimated 20% or lower rate of loss to follow-up is adequate to detect a 13% difference in failure rates between age groups with 80% power and a 95% confidence level. Only 1 surgical procedure was included per patient. If a patient underwent multiple meniscal repair operations, only the initial procedure was included. Patients undergoing meniscal repair with or without concurrent injuries or procedures were eligible. All root repairs were excluded from this analysis. Medical record data including all demographic, preoperative, intraoperative, and postoperative rehabilitation and complication data were included. Demographic data collected included age, height, weight, sports participation, past knee injuries, and surgical procedures. Operative time, presence of cartilage defects, ligament injury, and concomitant procedures were recorded as intraoperative data. Complications, reoperations, and failures were recorded for the postoperative period. Patient Selection, Surgical Technique, and Rehabilitation Isolated meniscal repairs were performed in patients with clinical symptoms of meniscal tears including joint-line pain and clicking or catching and the absence of significant concomitant degenerative joint disease (defined as joint space narrowing >50% on
weight-bearing flexion posteroanterior radiographs). At the time of arthroscopy, the decision for meniscal repair versus partial meniscectomy was based on surgeon assessment of tissue quality, tear vascularity, and surrounding cartilage status. In the setting of concomitant cruciate ligament reconstruction, meniscal repairs generally were performed on unstable tears and tears greater than 1.0 cm in length, although case-by-case discretion was used by the operating surgeon with consideration of tear pattern, tissue quality, and tear location including vascular zone and compartment (medial vs lateral). Small (<1.0 cm), stable peripheral vertical tears in the setting of anterior cruciate ligament (ACL) injury were frequently left in situ. Meniscal repair was performed with either an insideout or all-inside technique based on surgeon preference and tear location. Inside-out repairs were performed for posterior horn and body tears of the lateral or medial meniscus using standard posterolateral and posteromedial approaches, respectively, with double-loaded, straight, flexible needles with No. 2-0 braided suture (Smith & Nephew, Memphis, TN). All-inside repairs were performed with Fast-Fix implants or Fast-Fix 360 implants (Smith & Nephew). As an adjunct procedure, rasping of the meniscus and adjacent synovium was performed at the time of the repair to stimulate healing. After meniscal repair, all patients were kept noneweight bearing for 4 weeks with no knee flexion beyond 90 for 8 weeks. No resistive hamstring exercises were used for 8 weeks. In the setting of isolated meniscal repairs, jogging was allowed as early as week 10 and a return to sports was allowed as early as week 16. Aside from the initial weight-bearing and range-ofmotion restrictions to protect the meniscal repair, in patients who underwent meniscal repair and concomitant ACL reconstruction, rehabilitation was performed according to the Multicenter Orthopaedic Outcomes Network (MOON) protocol.21 Outcome Measures For the purpose of this study, meniscal repair failure was defined as a recurrent or unhealed tear on repeated magnetic resonance imaging (MRI), subsequent meniscectomy, no evidence of healing on repeated arthroscopy, revision meniscal repair, or total knee arthroplasty performed on the original surgically repaired meniscus. All patients who had recurrent or persistent symptoms of meniscal tear underwent a repeated knee MRI scan followed by diagnostic arthroscopy if the MRI findings were suggestive of an unhealed repair or a subsequent meniscal tear at a site other than the repair. In this series, there were no patients with persistent symptoms and abnormal repeated MRI scans who declined repeated arthroscopy. Manipulation under anesthesia, lysis of adhesions, and repeated arthroscopy with evidence of a healed meniscal repair were not considered failures. Meniscal
3
AGE AND RISK OF MENISCAL REPAIR FAILURE Table 1. Descriptive Statistics All Patients (N ¼ 221) Demographic characteristics Age, mean (SD), yr BMI, mean (SD) Sex, n Male Female Tear zone, n Vascular Avascular extension Avascular Tear pattern, n Longitudinal Radial Horizontal Complex Tear length, mean (SD), cm Time from injury to surgery, mean (SD), d Previous ipsilateral knee surgery, n Ligament status, n ACL injury PCL injury MCL or LCL injury Multiligament injury Grade 3-4 chondral defects present, n Concomitant procedure, n Repair technique, n All inside Inside out Meniscus repaired, n Medial Lateral Both No. of implants, mean (SD) Time to failure, mean (SD), mo
30.4 (11.7) 26.7 (9.5)
Patients Aged 40 yr (n ¼ 56) 47.2 (5.3) 28.8 (6.6)
Patients Aged <40 yr (n ¼ 165) 24.7 (6.7) 27.3 (5.9)
121 (54.8%) 100 (45.2%)
28 (50.0%) 28 (50.0%)
93 (56.4%) 72 (43.6%)
139 (62.9%) 81 (36.7%) 1 (0.5%)
28 (50.0%) 12 (21.4%) 0 (0.0%)
79 (47.9%) 42 (25.5%) 1 (0.6%)
P Value <.001 .18 .31
.44
.01 165 27 22 7 1.45 86 37
(74.7%) (12.2%) (10.0%) (3.2%) (0.51) (97) (16.7%)
38 6 11 1 1.23 94 11
(67.8%) (10.7%) (19.6%) (1.8%) (0.51) (102) (19.6%)
127 21 11 6 1.49 83 26
(77.0%) (12.7%) (6.7%) (3.6%) (0.51) (94) (15.8%)
163 1 7 2 29 212
(73.7%) (0.5%) (3.2%) (0.9%) (13.1%) (95.9%)
39 0 2 0 11 54
(69.6%) (0.0%) (3.4%) (0.0%) (19.6%) (96.4%)
124 1 5 2 18 158
(75.2%) (0.6%) (3.0%) (1.2%) (10.9%) (95.8%)
170 (76.9%) 51 (23.1%)
52 (92.9%) 4 (7.1%)
.06 .60 .73 .06
.11 >.99 .02
118 (71.5%) 47 (28.5%) .21
142 53 26 4.4 24.8
(64.3%) (24.0%) (11.8%) (3.3) (21.4)
38 15 3 3.5 16.9
(67.9%) (26.8%) (5.4%) (2.5) (10.2)
104 38 23 4.7 28.5
(63.0%) (23.0%) (14.0%) (3.4) (23.3)
.14 .04
ACL, anterior cruciate ligament; BMI, body mass index; LCL, lateral collateral ligament; MCL, medial collateral ligament; PCL, posterior cruciate ligament; SD, standard deviation.
surgery on the contralateral meniscus was not considered a failure. Cartilage defects that were determined to be Outerbridge grade 3 or 4 lesions at the time of arthroscopy were noted.22 Cartilage defects noted at the time of meniscal repair were treated with either debridement, chondroplasty, or microfracture. For the purpose of this study, meniscal tears were defined both by pattern (vertical or longitudinal, horizontal cleavage, radial, or complex pattern) and by vascular zone (vascular, vascular with avascular extension, or avascular). Information on tear length and number of sutures used was included as well. Root tears were excluded. Statistical Analysis All statistical analyses were performed using a standard software package (JMP Pro, version 12.2.0; SAS Institute, Cary, NC). Descriptive statistics were first generated for the entire data set. Simple logistic regression models were created for all potential predictors to assess for an unadjusted association with meniscal repair failure. A
multivariate logistic regression model was then created to assess the independent association between age of 40 years or older and meniscal repair failure status. Potential covariates included age, weight, sex, ACL status, number of implants used, tear pattern, tear vascularity, and chondral status at the time of the procedure. A backward selection was used with an exit criterion of a greater than .05. If there was evidence of confounding as defined as a change in the estimated effect size of greater than 15%23,24 for age of 40 years or older, the covariate was kept in the model regardless of the a level.
Results Descriptive Statistics Among the 276 eligible patients, 221 were successfully contacted for follow-up (80% follow-up [221 of 276 patients]).The sample population consisted of 100 men and 121 women at more than 5 years’ mean follow-up (mean, 64.1 months; standard deviation
4
S. POLAND ET AL.
Table 2. Univariate Risk Factors for Meniscal Repair Failure Individual Risk Factor Age 40 yr <40 yr Sex Female Male BMI Tear location Vascular zone Avascular extension Tear pattern Vertical Horizontal Radial Tear length Time from injury to surgery Previous ipsilateral knee surgery ACL injury Grade 3-4 chondral defects Repair technique All inside Inside out Side of repair Medial Lateral Both No. of implants Use of 5 implants (yes)
*
Unadjusted OR (95% CI)
P Value
Referent 1.24 (0.57-2.70)
NA .57
Referent 1.04 (0.54-2.01) Per 1.0: 1.03 (0.95,1.10)
NA .90 .48
Referent 0.87 (0.31-2.46)
NA .80
Referent 3.34 (0.44-25.2) 1.48 (0.41-5.28) Per 1.0 cm: 0.91 (0.20-4.14) Per 1.0 mo: 1.11 (0.95-1.30)
NA .24 .55 .91 .18
1.11 (0.46-2.60)
.84
0.94 (0.44-1.98) 0.61 (0.32-1.19)
.88 .15
Referent 0.66 (0.29-1.55)
NA .33
Referent 1.50 (0.64-3.53) 0.94 (0.20-4.46) Per 1.0 implant: 1.05 (0.95-1.15) 1.11 (0.53-2.30)
.35 .94 .38 .79
ACL, anterior cruciate ligament; BMI, body mass index; CI, confidence interval; NA, not applicable; OR, odds ratio. *Likelihood ratio c-square test.
[SD], 33.7 months). Of the 221 patients, 56 (25.3%) were aged 40 years or older (mean, 47.2 years; SD, 5.3 years) and 165 (74.7%) were younger than 40 years (mean, 24.7 years; SD, 6.7 years) at the time of surgery. Overall, 173 patients (78.2%) had a concurrent ligamentous injury along with their meniscal tear. A total of 163 (73.7%) had isolated ACL tears, 7 (3.2%) had isolated medial or lateral collateral ligament tears, 1 (0.5%) had an isolated posterior cruciate ligament tear, and 2 (0.9%) had multiligament injuries (Table 1). The mean meniscal tear length was 1.45 cm (SD, 0.51 cm), and the mean time from injury to surgery was 86 days (SD, 97 days). Of the meniscal tears, 62.9% (139 of 221) were entirely within the vascular zone, 36.7% (81 of 221) had an avascular extension component, and 0.5% (1 of 221) were avascular. A total of 76.9% of the meniscal tears (170 of 221) were repaired with the all-inside technique, whereas the other 23.1% (51 of 221) were repaired with the inside-out technique. Meniscal Repair Failures and Complications The overall meniscal repair failure rate was 20% (45 of 221 patients) (Table 1). Of the failures, 27 underwent
a subsequent meniscectomy of the repaired meniscus, 10 underwent a repeated meniscal repair, 6 underwent an ACL reconstruction along with a subsequent meniscectomy or meniscal repair, and 2 underwent an unspecified meniscal procedure. Two complications occurred, both in the group aged 40 years or older, consisting of 1 reported deep venous thrombosis and 1 emergency department visit for uncontrolled pain postoperatively. Among patients younger than 40 years, the failure rate was 21% (35 of 165) versus 18% (10 of 56) in the group aged 40 years or older. Of the 45 patients who were considered to have failures at follow-up, the mean time to failure was 24.8 months (SD, 21.4 months). Unadjusted Risk Factors for Meniscal Repair Failure In the univariate analysis, no variables were significantly associated with an increased risk of meniscal repair failure (Table 2). The mean time to failure was significantly shorter in patients aged 40 years or older (mean, 16.9 months; SD, 10.2 months) versus patients younger than 40 years (mean, 28.5 months; SD, 23.3 months) (P ¼ .04, t test with unequal variances), but the risk of failure at a mean of 5 years’ follow-up was equivalent (odds ratio [OR], 1.24; 95% confidence interval [CI], 0.57-2.70; P ¼ .57, likelihood ratio c-square test). The presence of cartilage defects at the time of arthroscopy (P ¼ .15), sex (P ¼ .90), tear pattern (P ¼ .80), tear length (P ¼ .91), time from injury to surgery (P ¼ .18), and prior knee surgery (P ¼ .84) were not predictive of meniscal repair failure. In addition, concurrent ACL injury (P ¼ .88), weight (P ¼ .73), and use of 5 or more implants during surgery (P ¼ .79) were not predictive of failure in the univariate analysis. Independent Relation Between Age and Risk of Failure After we controlled for confounding variables including weight, sex, ACL status, number of implants used, tear pattern, and chondral status at the time of the procedure, age of 40 years or older was not associated with an increased risk of meniscal repair failure (adjusted OR, 0.83; 95% CI, 0.36-1.81; P ¼ .65) (Table 3). Use of 5 or more sutures or meniscal implants showed a trend toward higher failure rates (adjusted OR, 1.79; 95% CI, 0.97-3.22; P ¼ .06) and was kept in the multivariate model because of evidence of moderate confounding of the effect of age group on failure rates (31% change in effect size when number of implants was removed from the model).
Discussion We found no significant difference in meniscal repair failure rates at 5-year follow-up between patients aged 40 years or older and patients younger than 40 years. Failures tended to occur earlier in the group aged
5
AGE AND RISK OF MENISCAL REPAIR FAILURE Table 3. Independent Risk Factors for Meniscal Repair Failure Independent Risk Factor Age 40 yr Use of 5 implants
Adjusted OR (95% CI)* 0.83 (0.36-1.81) 1.79 (0.97-3.22)
P Valuey .65 .06
CI, confidence interval; OR, odds ratio. *Adjustment performed as needed for sex, weight, tear pattern, repair technique, and concomitant injuries. y Likelihood ratio c-square test.
40 years or older than in the group younger than 40 years (mean, 16.9 months vs 28.5 months), although failure rates at final follow-up were similar. A trend toward higher failure rates was observed with use of 5 or more sutures or implants; no other independent risk factors were identified, including cartilage defects, prior knee surgery, weight, sex, tear pattern, and ACL status. The aforementioned failure rate is consistent with other studies, which have found that older patients have a similar, if not decreased, rate of meniscal repair failure.17-20 In fact, in a study by Lyman et al.17 in which age was treated as a continuous variable, older age was shown be a protective factor against meniscal repair failure, given that older patients tend to be less active and return to physically demanding activities at a lower and slower rate versus their younger counterparts. A decreased activity level among older patients may place fewer mechanical demands on the meniscal repair. In our study, meniscal repair failures in patients aged 40 years or older tended to occur earlier than failures in younger patients. Early failures could potentially be due to lack of healing at the repair site, although this cannot be proved definitively with the current study design. With increased age, the vascular supply of the meniscus is diminished.25 This hypovascular environment could potentially affect the initial healing ability of our meniscal repairs, contributing to a shorter time to failure in this older age cohort. In 2 studies that noted a higher meniscal repair failure risk with increased age, Eggli et al.16 similarly suggested that meniscal healing may be affected by decreased vascularity of the meniscus with increased age whereas Buseck and Noyes15 did not provide a possible explanation in their article. Limitations A limitation of this study is loss to follow-up of a portion of the original cohort. Although a minimum 80% follow-up was achieved, greater than 95% followup is considered ideal to reduce risk of bias. Contact attempts for each patient were made a minimum of 3 times by E-mail, phone, and mail until the patient was determined to be lost to follow-up. Functional and subjective outcome scores were not included in this
analysis. Although our study has one of the largest samples of patients aged 40 years or older in the published literature, it is still underpowered to detect small differences in failure risk between age groups. In addition, the results are based on a series of patients from 1 institution and may be less generalizable than multicenter study results. Most meniscal repairs in this study occurred in patients with concomitant ACL reconstructions; although ACL status was not associated with meniscal repair failure risk, it is possible that there are age-specific differences in isolated meniscal repair failure risk that were undetected in this study. Finally, asymptomatic patients did not routinely receive repeated knee MRI scans to assess meniscal repair healing; it is possible that some asymptomatic patients may have unhealed repairs that were not accounted for in this study’s definition of repair failure.
Conclusions Age of 40 years or older is not associated with an increased risk of meniscal repair failure at 5 years, although a shorter time to failure was noted in this age cohort.
References 1. Clayton RA, Court-Brown CM. The epidemiology of musculoskeletal tendinous and ligamentous injuries. Injury 2008;39:1338-1344. 2. Fithian DC, Kelly MA, Mow VC. Material properties and structure-function relationships in the menisci. Clin Orthop Relat Res 1990:19-31. 3. Roos H, Laurén M, Adalberth T, Roos EM, Jonsson K, Lohmander LS. Knee osteoarthritis after meniscectomy: Prevalence of radiographic changes after twenty-one years, compared with matched controls. Arthritis Rheum 1998;41:687-693. 4. Aagaard H, Verdonk R. Function of the normal meniscus and consequences of meniscal resection. Scand J Med Sci Sports 1999;9:134-140. 5. Lanzer WL, Komenda G. Changes in articular cartilage after meniscectomy. Clin Orthop Relat Res 1990:41-48. 6. Henry S, Mascarenhas R, Kowalchuk D, Forsythe B, Irrgang JJ, Harner CD. Medial meniscus tear morphology and chondral degeneration of the knee: Is there a relationship? Arthroscopy 2012;28:1124-1134.e2. 7. Fairbank TJ. Knee joint changes after meniscectomy. J Bone Joint Surg Br 1948;30:664-670. 8. Kim S, Bosque J, Meehan JP, Jamali A, Marder R. Increase in outpatient knee arthroscopy in the United States: A comparison of National Surveys of Ambulatory Surgery, 1996 and 2006. J Bone Joint Surg Am 2011;93:994-1000. 9. Abrams GD, Frank RM, Gupta AK, Harris JD, McCormick FM, Cole BJ. Trends in meniscus repair and meniscectomy in the United States, 2005-2011. Am J Sports Med 2013;41:2333-2339. 10. Paxton ES, Stock MV, Brophy RH. Meniscal repair versus partial meniscectomy: A systematic review comparing reoperation rates and clinical outcomes. Arthroscopy 2011;27:1275-1288.
6
S. POLAND ET AL.
11. Perdue PS, Hummer CD, Colosimo AJ, Heidt RS, Dormer SG. Meniscal repair: Outcomes and clinical follow-up. Arthroscopy 1996;12:694-698. 12. Pujol N, Tardy N, Boisrenoult P, Beaufils P. Long-term outcomes of all-inside meniscal repair. Knee Surg Sports Traumatol Arthrosc 2015;23:219-224. 13. Tuckman DV, Bravman JT, Lee SS, Rosen JE, Sherman OH. Outcomes of meniscal repair: Minimum of 2-year follow-up. Bull Hosp Jt Dis 2006;63:100-104. 14. Nepple JJ, Dunn WR, Wright RW. Meniscal repair outcomes at greater than five years: A systematic literature review and meta-analysis. J Bone Joint Surg Am 2012;94: 2222-2227. 15. Buseck MS, Noyes FR. Arthroscopic evaluation of meniscal repairs after anterior cruciate ligament reconstruction and immediate motion. Am J Sports Med 1991;19: 489-494. 16. Eggli S, Wegmüller H, Kosina J, Huckell C, Jakob RP. Long-term results of arthroscopic meniscal repair. An analysis of isolated tears. Am J Sports Med 1995;23: 715-720. 17. Lyman S, Hidaka C, Valdez AS, et al. Risk factors for meniscectomy after meniscal repair. Am J Sports Med 2013;41:2772-2778.
18. Barrett GR, Field MH, Treacy SH, Ruff CG. Clinical results of meniscus repair in patients 40 years and older. Arthroscopy 1998;14:824-829. 19. Noyes FR, Barber-Westin SD. Arthroscopic repair of meniscus tears extending into the avascular zone with or without anterior cruciate ligament reconstruction in patients 40 years of age and older. Arthroscopy 2000;16:822-829. 20. Steadman JR, Matheny LM, Singleton SB, et al. Meniscus suture repair: Minimum 10-year outcomes in patients younger than 40 years compared with patients 40 and older. Am J Sports Med 2015;43:2222-2227. 21. Wright RW, Haas AK, Anderson J, et al. Anterior cruciate ligament reconstruction rehabilitation: MOON guidelines. Sports Health 2015;7:239-243. 22. Outerbridge R. The etiology of chondromalacia patellae. J Bone Joint Surg Br 1961;43:752-757. 23. Maldonado G, Greenland S. Simulation study of confounder-selection strategies. Am J Epidemiol 1993;138: 923-936. 24. Mickey RM, Greenland S. The impact of confounder selection criteria on effect estimation. Am J Epidemiol 1989;129:125-137. 25. Gray JC. Neural and vascular anatomy of the menisci of the human knee. J Orthop Sports Phys Ther 1999;29:23-30.