Anterior cruciate ligament (ACL) rupture is one of the most common orthopaedic injuries in the United States, with an estimated 100,000 to 200,000 cases and over 100,000 surgical reconstructions annually.1 Although once thought to predominantly affect adults, recent studies have shown that ACL rupture is increasingly common among children and adolescents.1–4 In order to avoid early irreparable meniscal and chondral injuries secondary to instability,5–11 a higher number of pediatric patients now undergo ACL reconstruction (ACLR). These surgeries are also being performed at increasingly younger ages.1–3 The risk of subsequent graft failure in these patients may reach 25% or more, significantly higher than that for adults.5,12–17 In addition, rates of graft failure are not uniform across all pediatric patients, with incidence disproportionately high in subgroups such as females and soccer players.17,18
Although graft failure is perhaps the most commonly cited complication after ACLR, a recent study found that 28% of pediatric ACLR patients sustained a contralateral ACL rupture postoperatively,11 suggesting that a significant amount of postoperative risk for return to the operating room (RTOR) is due to problems other than graft failure. Other studies have detailed similar findings.11,13,19–22 Reported rates of any-cause RTOR after ACLR in pediatric patients range up to 38%.11,23 Anecdotally, this rate is significantly higher than the rate of graft failure alone due to instances of arthrofibrosis, infection, and subsequent meniscus injury. With most of the clinical and research focus specifically on the risk of re-tear, clinicians, patients, and families may benefit from a more comprehensive description of any-cause risk of RTOR after ACLR.
The goals of this study were to (1) describe the overall rate of RTOR among our patients at 1-, 2-, and 3-year post-ACLR; (2) estimate rates of RTOR by indication and determine their relative contributions to the overall risk of RTOR; and (3) compare rates of RTOR based upon whether meniscus surgery was performed at the index procedure.
After Institutional Review Board approval of this retrospective survivorship analysis, we identified potential subjects who underwent ACLR in our pediatric care network between May 2013 and December 2015 using the Current Procedural Terminology code for arthroscopic ACL reconstruction (29888). Patients were excluded in cases of revision ACLR, coexistent lower extremity fracture, congenital absence of the ACL, and planned staged knee reconstruction. Only primary ACLRs were included.
Baseline variables including age, sex, body mass index, and race were recorded. Clinic and operative notes were then reviewed to determine whether a meniscus repair or partial resection was performed with the index procedure. According to United States Centers for Disease Control and Prevention (CDC) guidelines, a patient was considered overweight if their body mass index for age percentile was ≥85. For each potential subject, we cataloged all Current Procedural Terminology codes assigned to the health record after the index procedure, noting cases where a subsequent operative procedure was related to either the index ACLR (ie, wound revision or infection) or a potential predilection for knee injury (ie, contralateral ACL tear). For subjects with more than one instance of RTOR, only the first instance was logged.
Cases of RTOR were grouped according to indication as shown below.
- Wound/hardware/infection issues—included cases of wound dehiscence, soft tissue infection, symptomatic implanted hardware, and dermatologic problems relating to incisions.
- Meniscus pathology—included subsequent procedures to address meniscus injuries. RTOR events with ACL graft failure and concomitant meniscus tear were considered under category 4, not category 2.
- Stiffness—included operative procedures for lysis of adhesions and other arthroscopic procedures to address range of motion deficits. Any manipulation under general anesthesia (MUA) was included in this category.
- ACL graft rupture—included cases of ipsilateral ACL graft failure.
- Contralateral ACL rupture—included cases of ACL rupture in the contralateral knee.
The Kaplan-Meier survival analysis was performed focusing on the primary outcome of any-cause RTOR. Patients without a RTOR event were censored after last known clinical follow-up. The 1-, 2-, and 3-year rates of RTOR were taken from the Kaplan-Meier analysis. Competing-risks regression was performed to predict rates of RTOR independently attributable to each cause category. These rates were then recorded at 1-, 2-, and 3-year after the index ACLR. The relative contributions of each category to any-cause RTOR were assessed by comparing individual rates of RTOR to the sum of the individual rates.
Cohorts were then defined based upon whether a meniscus procedure (repair and/or resection) was performed at index ACLR. The Kaplan-Meier analysis was again used to describe RTOR rates for each subgroup, and cohorts were compared using the log-rank tests.
A commercially available spreadsheet program was used to record patient variables in this study. All statistical analysis was performed using Stata 14.2 software (Statacorp, College Station, TX).
After inclusion and exclusion criteria were applied, 419 subjects underwent further abstraction and analysis. Median subject age was 15.4 years (interquartile range, 13.9 to 16.9). Other patient demographics for this study are shown in Table 1. Most ACLRs considered were transphyseal, with <20% involving a physeal-sparing approach. Survival analysis was limited to the first 3 postoperative years to ensure a sample size ≥50 subjects at each time point. At 3-year post-ACLR, the any-cause risk of RTOR was 16.5% (95% confidence interval, 12.4%-21.9%). By-cause RTOR rates were recorded from the regression analysis and are shown in Table 2. According to these results, ACL graft rupture represented approximately 43% of the overall risk of RTOR by 3-year post-ACLR, consistent with the finding that 47% of fist-time RTOR events in the study sample were attributed to ACL graft failure. This relationship is illustrated in Figure 1.
Because of concerns regarding the potential for varying risk of RTOR by patient age or maturity, patients were separated into 4 groups based upon age quartiles (0 to ≤13.9, >13.9 to ≤15.4, >15.4 to ≤16.9, >16.9), and subgroup Kaplan-Meier analysis was performed. This revealed by-group estimates for 3-year RTOR to be 16.5%, 20.7%, 14.0%, and 13.6%, respectively. Any difference between these rates was not statistically significant (log-rank test; P=0.559). In addition, differences in RTOR rates by sex, overweight status, and physeal-sparing (vs. transphyseal) reconstruction were not statistically significant (P=0.631, 0.788, 0.474, respectively).
RTOR category 1 included cases of wound breakdown, wound excision and revision, soft tissue incision and debridement for infection, and one case of symptomatic cortical button removal. These complications occurred primarily in the first 3 to 6 months after surgery, accounting for the majority of any-cause RTOR in the early postoperative phase. Category 2 included ipsilateral meniscus tears and a single contralateral meniscus tear. Again, this was included to provide a comprehensive perspective on RTOR after ACLR. Meniscus operations tended to occur ≥1 year after ACLR. Patients who returned to the OR for arthroscopy or MUA after failing to meet knee range of motion goals were included in category 3. These cases occurred most frequently around 9 to 12 months after ACLR, although the single case of MUA noted occurred between 3 and 4 months post-ACLR.
Cases of ACL graft failure first appeared in our study population around 6 months after surgery and rose dramatically around 9 to 12 months. Notably, almost half of re-tears (10/23) occurred before the patient’s clearance to return to full activity. Contralateral ACL ruptures tended to occur between 1 and 2 years after surgery.
Cohort-specific rates of any-cause RTOR are shown in Table 3. Patients who had a meniscus repair or resection had lower rates of 3-year RTOR than those who did not (P=0.043). Among patients without a meniscus procedure, 9.8% experienced initial RTOR due to graft failure compared with only 3.0% for those with a meniscus procedure.
Interest in complications following arthroscopic knee surgery developed in the 1980s with several studies reporting perioperative iatrogenic complication rates under 2% in adults.24–26 Over the years, such studies have found that the risk of graft failure in adult ACLR may range up to 7.9%.13 More recent investigations of ACLR outcomes have further examined rates of graft failure, reporting re-tear rates as high as 25% among general pediatric patients5,12,14–16 and 34% for female soccer players who return to soccer.17 However, the traditional research focus on graft failure may underestimate the likelihood of future knee problems for these young patients.
While the post-ACLR management paradigm conventionally frames graft failure as the primary risk after surgery, surgeons and families may be interested in any-cause and by-cause rates of RTOR after pediatric ACLR. In this study, we thus sought to explore more broadly the likelihood of a pediatric patient having further operative knee issues after ACLR. Accordingly, we included issues that may not formally be considered “complications” after reconstruction (ie, contralateral ACLR) as cases of RTOR. As we considered more than just ipsilateral graft failure as an outcome, we report a RTOR rate (approximately 1 in 6 patients) significantly higher than most cited complication rates. However, our rate of RTOR is still comparable with a recently published estimate of RTOR gleaned from a significantly smaller patient sample (24% at 10 y post-ACLR).23 In all, the high rate of RTOR we found is thought to reflect a comprehensive accounting of postreconstruction knee problems.
We believe that the high likelihood of RTOR in pediatric patients—and particularly the high risk of graft failure—is largely due to their difficulties with adherence to postoperative restrictions and the relatively high activity levels to which they attempt to return. This idea is supported by our finding that almost half of re-tears observed in our sample occurred before clearance for full activity. The added technical difficulty of ACLR in children and adolescents (vs. adults), especially in skeletally immature patients, should also be considered as a potential contributor to the increased rate of complication. While the smaller body of experience that surgeons generally have operating on children could inflate the rate of post-ACLR complications on a population basis, this phenomenon likely did not affect our analysis as the surgeons involved were dedicated pediatric sports medicine surgeons.
Although graft failure may be the most commonly reported cause for RTOR in prior literature, it represented less than half of the overall risk in this study population. We found no instances of postoperative embolism, compartment syndrome, or death, which are rare complications reported in the literature.27 Although we found a very small number of cases with concern for superficial infection, we found no subjects with frank iatrogenic knee sepsis. For the types of postoperative issues that we did encounter, we provided approximate timelines detailing when they occurred after surgery.
Unlike previous investigations, we compared ACLR outcomes based upon concomitant meniscal intervention. We found that patients who underwent ACLR with meniscal repair or resection had lower 3-year RTOR rates than those who did not. This is consistent with the findings of a previous study showing that meniscal repairs with concomitant ACLRs have a lower reoperation rate than isolated meniscal repairs, possibly due to synergism in ACL graft and meniscus healing.28 We also found that the cohort with a meniscus procedure experienced first-time RTOR for ACL graft failure at a lower rate than its complementary cohort. However, we could not statistically compare overall rates of graft failure between these groups because only first-time RTOR cases were logged. In all, these findings could potentially reflect decreased postreconstruction knee instability in patients with meniscus procedures leading to lower rates of RTOR for meniscus and ACL pathology. However, it is possible that early meniscal management simply minimizes the chances of propagation of minor meniscal damage existing at the time of the index procedure.
This study has several limitations. First, our findings detail only a single pediatric care center’s experience with ACLR, and thus may not be generalizable to all sites. We did not record subsequent RTOR after each subject’s first event, therefore competing-risks regression was used to estimate by-cause complication rates. As we did not have records for RTOR procedures performed outside of our health system, we concede that true rates of RTOR are potentially higher than estimated here. This study was not designed to examine overall rates of graft failure between patients with and without a meniscus procedure, although this represents interesting future work.
A major limitation of this study is that we could not directly compare rates of RTOR between various ACLR techniques, especially techniques used in skeletally immature children. While this study was performed at a pediatric care network that performs these procedures and many providers are interested in such a quantitative comparison, our center performs 4 to 5 times as many transphyseal reconstructions as physeal-sparing surgeries each year. Thus, a significantly larger cohort would have been needed to compare the various subgroups of interest. Further, because we did not perform skeletal age measurements from direct radiograph assessment, we could not assess the impact of skeletal age on the rate of RTOR. Because only a small minority of patients had formal hand/wrist films to quantify skeletal maturity, chronologic age, and sex were considered independent of radiographic maturity ratings.
This study is one of the first to focus on any-cause RTOR after ACLR. As the literature focuses almost exclusively on graft failure, this study provides a more comprehensive synopsis of outcomes for both providers and patients. This work may serve as an effective pilot study for new projects investigating ACLR outcomes. Larger prospective studies with longer follow-up are warranted to further investigate outcomes and complications after ACLR. Multicenter study designs may be used to increase the number of evaluable physeal-sparing reconstructions to allow for future comparative analysis of RTOR rate by reconstruction type.
1. Mall NA, Chalmers PN, Moric M, et al. Incidence and trends of anterior cruciate ligament
reconstruction in the United States. Am J Sports Med. 2014;42:2363–2370.
2. Buller LT, Best MJ, Baraga MG, et al. Trends in anterior cruciate ligament
reconstruction in the United States. Orthop J Sports Med. 2015;3:2325967114563664. 1–8.
3. Dodwell ER, Lamont LE, Green DW, et al. 20 years of pediatric anterior cruciate ligament
reconstruction in New York State. Am J Sports Med. 2014;42:675–680.
4. Milewski MD, Beck NA, Lawrence JT, et al. Anterior cruciate ligament
reconstruction in the young athlete: a treatment algorithm for the skeletally immature. Clin Sports Med. 2011;30:801–810.
5. Aichroth PM, Patel DV, Zorrilla P. The natural history and treatment of rupture of the anterior cruciate ligament
in children and adolescents. A prospective review. J Bone Joint Surg Br. 2002;84:38–41.
6. Dumont GD, Hogue GD, Padalecki JR, et al. Meniscal and chondral injuries associated with pediatric anterior cruciate ligament
tears: relationship of treatment time and patient-specific factors. Am J Sports Med. 2012;40:2128–2133.
7. Henry J, Chotel F, Chouteau J, et al. Rupture of the anterior cruciate ligament
in children: early reconstruction with open physes or delayed reconstruction to skeletal maturity? Knee Surg Sports Traumatol Arthrosc. 2009;17:748–755.
8. Lawrence JT, Argawal N, Ganley TJ. Degeneration of the knee joint in skeletally immature patients with a diagnosis of an anterior cruciate ligament
tear: is there harm in delay of treatment? Am J Sports Med. 2011;39:2582–2587.
9. Millett PJ, Willis AA, Warren RF. Associated injuries in pediatric and adolescent anterior cruciate ligament
tears: does a delay in treatment increase the risk of meniscal tear? Arthroscopy. 2002;18:955–959.
10. Ramski DE, Kanj WW, Franklin CC, et al. Anterior cruciate ligament
tears in children and adolescents: a meta-analysis of nonoperative versus operative treatment. Am J Sports Med. 2014;42:2769–2776.
11. Schmale GA, Kweon C, Larson RV, et al. High satisfaction yet decreased activity 4 years after transphyseal ACL reconstruction. Clin Orthop Relat Res. 2014;472:2168–2174.
12. Koizumi H, Kimura M, Kamimura T, et al. The outcomes after anterior cruciate ligament
reconstruction in adolescents with open physes. Knee Surg Sports Traumatol Arthrosc. 2013;21:950–956.
13. Magnussen RA, Meschbach NT, Kaeding CC, et al. ACL graft and contralateral ACL tear risk within ten years following reconstruction: a systematic review. JBJS Rev. 2015;3:1–8.
14. Nelson IR, Chen J, Love R, et al. A comparison of revision and rerupture rates of ACL reconstruction between autografts and allografts in the skeletally immature. Knee Surg Sports Traumatol Arthrosc. 2016;24:773–779.
15. Peterson DC, Ayeni OR. Pediatric anterior cruciate ligament
reconstruction outcomes. Curr Rev Musculoskelet Med. 2016;9:339–347.
16. Sanders TL, Pareek A, Hewett TE, et al. Long-term rate of graft failure after ACL reconstruction: a geographic population cohort analysis. Knee Surg Sports Traumatol Arthrosc. 2017;25:222–228.
17. Allen MM, Pareek A, Krych AJ, et al. Are female soccer players at an increased risk of second anterior cruciate ligament
injury compared with their athletic peers? Am J Sports Med. 2016;44:2492–2498.
18. Shea KG, Pfeiffer R, Wang JH, et al. Anterior cruciate ligament
injury in pediatric and adolescent soccer players: an analysis of insurance data. J Pediatr Orthop. 2004;24:623–628.
19. Ho B, Edmonds EW, Chambers HG, et al. Risk factors for early ACL reconstruction failure in pediatric and adolescent patients: a review of 561 cases. J Pediatr Orthop. 2016. [Epub ahead of print].
20. Magnussen RA, Reinke EK, Huston LJ, et al. Factors associated with high-grade lachman, pivot shift, and anterior drawer at the time of anterior cruciate ligament
reconstruction. Arthroscopy. 2016;32:1080–1085.
21. Paterno MV, Rauh MJ, Schmitt LC, et al. Incidence of second ACL injuries 2 years after primary ACL reconstruction and return to sport. Am J Sports Med. 2014;42:1567–1573.
22. Webster KE, Feller JA, Leigh WB, et al. Younger patients are at increased risk for graft rupture and contralateral injury after anterior cruciate ligament
reconstruction. Am J Sports Med. 2014;42:641–647.
23. Reid D, Leigh W, Wilkins S, et al. A 10-year retrospective review of functional outcomes of adolescent anterior cruciate ligament
reconstruction. J Pediatr Orthop. 2017;37:133–137.
24. DeLee JC. Complications of arthroscopy and arthroscopic surgery: results of a national survey. Committee on Complications of Arthroscopy Association of North America. Arthroscopy. 1985;1:214–220.
25. Lindenbaum BL. Complications of knee joint arthroscopy. Clin Orthop Relat Res. 1981;160:158.
26. Small NC. Complications in arthroscopic surgery performed by experienced arthroscopists. Arthroscopy. 1988;4:215–221.
27. Ferrari D, Lopes TJ, Franca PF, et al. Outpatient versus inpatient anterior cruciate ligament
reconstruction: a systematic review with meta-analysis. Knee. 2017;24:197–206.
28. 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.