Abstract
Background:
Posterior cruciate ligament (PCL) avulsion fractures pose unique technical challenges for surgical fixation in the pediatric and adolescent population. Both arthroscopic and open management have been pursued; nonetheless, the literature is devoid of comparative studies elucidating factors for technique selection or relative outcomes.
Purpose:
To compare the demographic characteristics, presenting features, clinical course, return to sports (RTS), patient-reported outcomes (PROs), and complications associated with open and arthroscopic fixation cohorts of pediatric patients who sustained an isolated tibial-sided PCL avulsion fracture.
Study Design:
Cohort study; Level of evidence, 3.
Methods:
A medical record review was performed for all patients aged 8 to 19 years who underwent operative fixation of a tibial-sided PCL avulsion fracture between 2003 and 2021 at a single pediatric tertiary referral center. Patients were excluded if they sustained a femoral-sided avulsion fracture, underwent a PCL reconstruction, or presented with multiligamentous knee injury requiring surgical intervention for >1 ligament. All patients were sent 3 PRO instruments validated for children: Pediatric International Knee Documentation Committee, Knee injury and Osteoarthritis Outcome Score-Child, and Hospital for Special Surgery Functional Activity Brief Scale.
Results:
A total of 20 patients met the inclusion criteria, of whom 10 patients were treated with open fixation and 10 with arthroscopic fixation. Also, 15 (75%) patients (7 open, 8 arthroscopic) returned PROs. No differences were detected in demographic characteristics, RTS, or PROs between cohorts. One patient in the open cohort developed a deep-tissue infection in the popliteal fossa, requiring open irrigation and debridement. One patient in the arthroscopic cohort eventually required PCL reconstruction for fracture nonunion associated with persistent instability.
Conclusion:
No differences in outcomes were detected in this study between cohorts of pediatric patients undergoing open and arthroscopic fixation of tibial-sided PCL avulsion injuries. Regardless of surgical technique, most patients healed without complications and returned to high activity levels. Given the postoperative complications seen in the arthroscopic cohort, further research is needed to investigate the potential biomechanical superiority of open technique fixation constructs.
The posterior cruciate ligament (PCL) is the primary restraint against posterior tibial translation at the knee joint. The PCL extends from the anterolateral aspect of the medial femoral condyle to the posterior proximal tibia, just distal to the joint line. The typical mechanism of injury that causes a PCL tear is either direct blunt force trauma to the anterior proximal tibia while the knee is flexed (ie, “dashboard injury”) or an indirect hyperflexion or hyperextension injury. However, knee injury resulting in an isolated PCL tear is rare, with the true incidence remaining largely unknown. 6 PCL tears are estimated to make up 3% to 23% of all knee injuries in adults, with significant variability existing likely due to the difficulty of identifying these injuries and the subtlety with which they present.6,7,10,14
In the pediatric and adolescent population, PCL injury is even less common, and isolated midsubstance tears present less frequently than do tears associated with bony avulsion fractures.6,10 While evidence-based literature is limited, a PCL tear associated with any displaced fracture and/or avulsion at the tibial insertion is generally considered a clear indication for surgery based on the success of fracture fixation of these fragments and the lack of long-term data on the natural history of nonoperative treatment in this age group. However, there are concerns regarding chronic knee instability and potential insidious onset of patellofemoral and medial compartment arthritis, which has been described for PCL-deficient knees in older populations. 3 The optimal surgical approach and technique for repair of a tibial-sided PCL avulsion fracture have yet to be elucidated. Thus far, these injuries have been treated according to individual surgeon expertise and comfort level. While arthroscopic techniques have been documented in the literature,1,3,6-10,14 the attachment site of the PCL to the posterior intercondylar fossa of the tibia is a notably difficult location to access arthroscopically.8,9,14 Despite the risks associated with open repair when compared with arthroscopy, as well as the anatomy relevant to the posterior approach to the knee, it has been hypothesized that open reduction of these avulsed bony fragments may provide biomechanically superior fixation and better long-term knee function. Such outcomes may be particularly important for the pediatric and adolescent population, especially in cases of competitive athletes with high-level return to sports (RTS) goals. Therefore, this study aimed to compare the demographic characteristics, presenting features, clinical course, complications, RTS, patient-reported functional outcomes (PROs), and secondary or revision surgery between open and arthroscopic cohorts of pediatric patients who underwent fixation of an isolated tibial-sided PCL avulsion fracture.
Methods
After institutional review board approval, a retrospective medical record review was performed for all patients aged 8 and 19 years who underwent surgical repair of a tibial-sided PCL avulsion fracture between 2003 and 2021. Patients were treated by 1 of 7 dual fellowship-trained pediatric orthopaedic sports medicine surgeons (Y.M.Y., D.E.K., M.S.K., B.E.H.). In terms of technique utilized, treatment decisions were surgeon-dependent, although 1 of the 7 surgeons performed procedures in each of the 2 technique-based cohorts. Patients were excluded if they underwent PCL reconstruction, had a femoral-sided avulsion fracture, or presented with multiligamentous knee injury requiring operative intervention.
Open Surgical Technique
After smooth induction of general anesthesia and intubation, the patient was brought from a supine position to a prone position, with special attention taken to adequately pad all down surfaces and ensure that genitalia were free of any pressure points. The operative knee was bumped up using several blankets to allow for lateral fluoroscopic images to be higher than the contralateral leg. The operative extremity was then prepared and draped in a routine sterile fashion. A standard posteromedial approach to the knee involved a 5 to 6–cm incision beginning at the border of the posteromedial hamstring tendons and extending distally to the proximal medial gastrocnemius (Figure 1). Careful deep dissection involved identifying the saphenous nerve and retracting it medially throughout the case. The semimembranosus was identified, and its fascial sheath incised. The medial head of the gastrocnemius was elevated and retracted laterally, allowing for identification of the posterior tibial plateau and its connections to the semimembranosus fascia and the origin of the popliteus. Both connections were incised, and the popliteus was elevated slightly to better visualize the underlying fracture fragment and fracture bed. The fracture fragment was then elevated, and the fracture bed irrigated and curettaged to optimize the healing surface of the anatomic reduction. Provisional fixation of the fragment was achieved using a 0.045 Kirshner wire, and optimal position was confirmed with fluoroscopic images showing the guidewire between the proximal tibial physis and subchondral bone of the tibial plateau. The guidewire from the 4.0 cannulated screw set was used. Screw fixation was, in some cases, augmented with suture anchor fixation, typically using nonabsorbable high-strength braided No. 2 suture in a Mason-Allen suture construct.
(A) Posteromedial approach to the left knee demonstrating the intracapsular view of the tibial PCL ligamentous footprint intact on the bony fragment, which is displaced from its adjacent bony bed. (B) Postfixation photo shows a screw head and washer on the avulsion fracture fragment, having been reduced anatomically into the fracture bed. (C) Postfixation radiograph shows the physeal-respecting, all-epiphyseal position of the partially threaded cannulated screw and washer construct anatomically fixing the fracture fragment in place. PCL, posterior cruciate ligament.
Arthroscopic Surgical Technique
The patient was placed on the operating room (OR) table in a supine position, with adequate padding placed on all down surfaces. A circumferential leg holder was attached to the table on the operative side, and the operative leg was prepared and draped in routine sterile fashion. Standard anteromedial and anterolateral arthroscopic portals were established, and a diagnostic arthroscopy was performed (Figure 2). A posteromedial portal was then established to better access the PCL and the avulsed bony fragment, as well as the fracture bed. A combination of arthroscopic electrocautery, motorized shaver, and periosteal elevator was used to debride the fracture bed to optimize anatomic reduction. Depending on the anatomy of the patient and the fracture, either an anterior cruciate ligament or PCL guide system can be used to create 2 tibial tunnels on either side of the fracture bed. Suture passers were then introduced through the tunnels using a 2.4-mm guidewire, and two 2-0 high-strength, nonabsorbable braided No. 2 sutures were passed through the base of the PCL just adjacent to the avulsed bony fragment and then fed through the tunnels. An anterior drawer maneuver is then applied, during which time the sutures are tied over the anterior cortical bone bridge between the 2 tibial tunnels (Figure 3). The tension on the fragment was confirmed arthroscopically, and the knee was then taken through a full range of motion with posterior drawer and sag reexamined. Arthroscopic portal sites were then closed in routine fashion.
(A) Arthroscopic view of the PCL from the anterolateral portal, with a relatively normal appearance despite the tibial-sided avulsion injury. (B) Assessing the tension of the PCL with the arthroscopic probe from the anteromedial portal reveals gross laxity of the ligament, stemming from the tibial-sided avulsion. (C) View of the elevated tibial-sided avulsion of the PCL after suture passage through the most distal 8 mm of the ligament with an arthroscopic suture passage device, and shuttling of the sutures provisionally into the posterolateral portal. (D) Tension on the sutures confirms good control of the entire PCL ligamentous complex with the previously passed high-strength No. 2 braided sutures. (E) View from the posterolateral portal shows the 2.4-mm guidewire in the far posterior aspect of the ligamentous footprint of the PCL on the tibia. (F) The 2.4-mm guidewire has been removed, and a looped suture passing device has been inserted up the tibial tunnel previously made by the 2.4-mm guidewire. The PCL sutures have been shuttled through the suture passing device. (G) The PCL sutures pulled almost completely through the tibial tunnel, with the PCL almost completely tensioned to the far posterior aspect of the tibial PCL footprint. (H) The PCL sutures have been maximally tensioned and impacted with a knotless suture anchor into the anterior tibial metaphysis just distal to a 2.4-mm tibial tunnel. The PCL appears to have normal anatomic tension and contour when viewed from the posterolateral portal. Examination under anesthesia reveals a normal posterior drawer test with a solid endpoint. (I) The PCL, when viewed from the anterolateral portal, appears taut and normally tensioned after suture fixation. PCL, posterior cruciate ligament.
(A) Injury radiograph. (B) The early postoperative radiograph demonstrates anatomic reduction, with PCL repair sutures having been fed through a transtibial tunnel and tied over an endobutton on the anteromedial tibial cortex. (C) The 6-month postoperative radiograph shows advanced healing of the PCL fracture fragment. PCL, posterior cruciate ligament.
The period of data collection for the study, both retrospective medical record review and patient outreach for PROs, was between December 2021 and December 2022. After determining the study cohorts, charts of included patients were reviewed for demographic characteristics, injury characteristics, surgical details, postoperative activity progression, complications (eg, infection, failure of fixation construct, postoperative range of motion restriction, and growth disturbance), and any subsequent knee surgeries. Patient outreach was then performed. All patients were sent questionnaires containing 3 PRO instruments validated for children. Outreach was done via emails containing secure links to the outcome questionnaire within a study-specific REDCap database, physical mail, and phone calls. Along with the PRO questionnaire, all patients were provided with a cover letter informing them of the nature of the study, associated risks, and study materials. Also, they were informed that their participation in the study was entirely voluntary. Patients were also allowed to opt out of both the PRO survey and the retrospective data collection.
The PRO instruments used were the Pediatric International Knee Documentation Committee subjective knee form (Pedi-IKDC), the Knee Injury and Osteoarthritis Outcomes Score for children (KOOS-Child), and the Hospital for Special Surgery Functional Activity Brief Scale (HSS-Pedi-FABS). The Pedi-IKDC is a validated knee-specific measure of patient-reported symptoms, function, and sports activity that has been modified from its original form to enhance generalizability for the broad adult orthopaedic population. 11 It is a 13-item questionnaire that is scored to yield a numeric value between 0 and 100, with higher scores representing higher levels of function/activity and less negative symptoms. The KOOS-Child is a 5-dimensional scale evaluating pain (KC-P), symptoms (KC-S), difficulty during activities of daily living (KC-ADL), function in sports/play (KC-S/P), and knee-related quality of life (KC-QoL). While originally designed to measure knee function in adult joint injury and/or degenerative disease, it has been modified to evaluate outcomes of a wide range of traumatic knee injuries in children. 13 Each subscale within the KOOS-Child is scored separately, after which the 5 scores are combined and converted to a single numeric score on a scale of 0 to 100, where a higher score represents better knee function. The HSS-Pedi-FABS is an 8-item questionnaire that more specifically quantifies the level of athletic and sport-specific activity.4,5 The questionnaire is scored to yield a numeric value between 0 and 30, with higher scores representing higher levels of athletic performance. Based on previous research utilizing these same outcomes, the mean ± standard deviation has been reported as 59 ± 22 for the pedi-IKDC 11 and 21.12 ± 7.51 for the HSS Pedi-FABS. 4 Power analysis determined that samples of 17 patients per group would provide 80% power to detect an effect size of 1 across groups using a 2-sample t test with the alpha set to 5%.
Statistical Analysis
Continuous numerical data were summarized using the median and interquartile range (IQR). Categorical data were summarized as the number of patients and the percentage frequency per category. The Mann-Whitney U test, the Pearson chi-square test, and the Fisher exact test were used as appropriate to compare demographic characteristics and outcomes across cohorts, with statistical significance set at P < .05. Any patient with a missing data value for any given variable was excluded from the given analysis. All statistical analyses were performed with JMP Pro 17.0.0 statistics software. Power analysis was performed using G*Power 3.1.9.6 software.
Results
Over the study period, 20 patients met the inclusion criteria. The median [IQR] patient age at the time of surgery was 13.5 years [12.5-14 years], and 13 (65%) patients were male. Most patients were skeletally immature, with 16 (80%) patients demonstrating open distal femoral/proximal tibial physes, 3 (15%) patients with closing physes, and 1 (5%) patient with closed physes. The most common mechanism of injury was a direct-blow contact injury seen in 12 (60%) patients, and the activities observed at the time of injury were sports, bicycle/scooter accident, motor vehicle accidents (MVA), fall from an elevated height (>3 ft), fall from standing height, and playground/child horseplay. All except 2 patients had completely isolated tibial-sided PCL avulsion fractures. One patient sustained a PCL injury as well as a proximal diaphyseal oblique femur fracture in an MVA, and the femur fracture was treated with rigid, trochanteric entry intramedullary nailing 1 month before her arthroscopic PCL repair. Another arthroscopic patient was noted to have lateral meniscus fraying intraoperatively. Still, there were no signs of full-thickness or meniscus root tear; thus, surgical intervention on the meniscus was not indicated. The median size of the avulsed bony fragment in the superior-inferior, medial-lateral, and anterior-posterior dimensions was 14 mm, 16 mm, and 4.5 mm, respectively. The median size in 3 dimensions of the avulsed bony fragment was 1.02 cm3. The median [IQR] time to full weightbearing (WB), time to running, and time to RTS were 6.3 [6-7] weeks, 3.1 [3-4.5] months, and 6.2 [5.7-8.3] months, respectively (Table 1).
Demographic Characteristics of the Cohort a
BMI, body mass index; IQR, interquartile range. Dashes indicate value is not applicable.
Ten patients underwent open reduction and internal fixation of the avulsed tibial-sided fragment, and 10 underwent arthroscopic fixation and repair. There were no instances of arthroscopic cases converted to open cases, or vice versa. The median [IQR] length of surgery for the open cohort was shorter, at 1.9 hours [1.7-2.2 hours], compared with that of the arthroscopic cohort, at 2.5 hours [1.4-3.2]. Differences in age, skeletal maturity, injury mechanism or activity, size of bony fragment, time to surgery, time to full WB, time to running, and RTS between open and arthroscopic cohorts are presented in Table 2.
Open Versus Arthroscopic Data and Retrospective Clinical Outcomes a
BMI, body mass index; IQR, interquartile range. Dashes indicate value is not applicable.
PRO survey responses were obtained from 15 (75%) patients: 7 from the open cohort and 8 from the arthroscopic cohort. Overall, for both cohorts, the median [IQR] Pedi-IKDC and HSS-Pedi-FABS scores were 96.7 [94.6-100] and 19 [16.5-21.8], respectively. KOOS-Child subscore medians were as follows: KC-P: 98.6; KC-S: 73.2; KC-ADLs: 100; KC-S/P: 97.5; and KC-QoL: 96.9. Survey responses by surgical cohort are presented in Table 3.
Open Versus Arthroscopic Patient-Reported Functional Outcomes a
HSS-Pedi-FABS, Hospital for Special Surgery Functional Activity Brief Scale; IQR, interquartile range; KOOS, Knee injury and Osteoarthritis Outcome Score; Pedi-IKDC, pediatric international knee documentation committee; PRO, patient-reported outcome.
Several complications were observed in both treatment groups. In the open cohort, 1 patient developed a deep-tissue infection in the popliteal fossa associated with subacute osteomyelitis. The infection occurred 1 year after the initial repair. He required return to the OR for open irrigation and debridement, and removal of all implants. His OR cultures were positive for methicillin-sensitive Staph Aureus, Staph Epidermidis, and Staph hominis, and he was further treated with a 12-week course of antibiotics (3 weeks intravenous followed by 9 weeks oral). PRO scores for this patient 1 year after the second procedure were as follows: Pedi-IKDC: 89; KC-P: 100; KC-S: 71; KC-ADLs: 100; KC-S/P- 85; and KC-QoL- 69. A second patient in the open cohort developed a superficial wound/suture infection that was treated with a short course of oral cephalexin.
In the arthroscopic cohort, 1 patient sustained a failed repair in the form of recurrent avulsion of the bony fragment 6 months after the initial repair. This caused her persistent knee instability and chronic PCL insufficiency requiring 3 additional surgeries: arthroscopic removal of loose body, lateral meniscal repair, and ultimately PCL reconstruction with an allograft. PRO scores for this patient 6 years after the most recent procedure were as follows: Pedi-IKDC- 59; KC-P: 67; KC-S: 54; KC-ADLs: 74; KC-S/P: 40; and KC-QoL: 50. A second patient in the arthroscopic cohort sustained a partially disrupted repair seen on plain radiographs 3 months after surgery, with an unclear mechanism of injury, given that the patient denied a fall or noncompliance in the early postoperative period. He had consistent activity-related knee pain, subjective feelings of instability, and a grade 2 posterior drawer test, all of which were attributed to the disrupted repair for as long as 2 years after surgery. At this point, he was given another physical therapy prescription and was subsequently lost to follow-up and did not complete the longer-term PRO prospective outreach. No additional complications were seen in either cohort.
Discussion
Although rare, PCL injury in the pediatric and adolescent population has been reasonably well described, particularly PCL avulsion fractures. Furthermore, PCL disruption involving bony, chondro-epiphyseal, or purely ligamentous avulsion at the tibial insertion has emerged, based on case reports to date, as an indication for surgery in this age group. Both open and arthroscopic approaches have been documented in the literature, and treatment to this point has largely been dictated by individual surgeon preference, as no study has directly compared outcomes of the 2 techniques. The purpose of this study was, therefore, to compare both clinical outcomes and PROs between open and arthroscopic cohorts of patients who sustained tibial-sided PCL avulsion injuries.
Ultimately, the study detected no differences in any of the clinical variables observed (see Table 2). The arthroscopic cohort had a longer median length of surgery by >30 minutes, with a much wider IQR compared with the open cohort, but ultimately, the median time to full WB, running, and RTS was similar in both groups.
The study also detected no differences in any PRO scores between the open and arthroscopic patients (Table 3). All median scores were very similar between cohorts across the board, and nearly all scores were equal to or greater than normative values for the healthy pediatric population.2,5,12 The only deviations from this trend were that both cohorts scored a fair amount lower than the healthy normative population in the KC-S, and the arthroscopic cohort was below this normative value in KC-QoL. While there are many more nuances and details to examine when comparing surgical techniques, these QoL questions are designed to focus on the crux of patient concerns about how they will ultimately recover from their operation. This discrepancy reveals an important viewpoint on the long-term patient experience with both approaches, and further investigation with larger patient numbers is needed, given the nature of this subscale.
The major complications experienced in both groups also represent a critical difference in long-term outcomes between open and arthroscopic cases, particularly when looking at the PROs submitted by the patients. The patient from the arthroscopic group who sustained recurrent avulsion of the bony fragment was ultimately considered a biomechanical failure of the initial repair, resulting in years of knee instability and chronic PCL insufficiency that would later require PCL reconstruction. Her PRO scores were well below not just the healthy population normative values, but the arthroscopic group median for all outcome instruments as well. In addition, while his symptoms were much less severe and he did not require additional surgery, the patient with persistent postoperative activity-related pain and partially disrupted repair seen on radiograph 3 months after surgery could also be considered a failure of arthroscopic repair. Conversely, there were no biomechanical failures in the open cohort. Despite the severity and risks associated with a case of osteomyelitis and deep-tissue abscess, the mechanical stability of this patient's repair remained intact, and all his PRO scores except KC-QoL fell within the IQR for the cohort, with his KC-P, KC-S, and KC-ADL scores each being equal to the open cohort median value. The other patient had a very mild superficial suture infection treated with <1 week of oral antibiotics and had all PRO scores equal to the open cohort median values. While low patient numbers for this investigation prevent any meaningful statistical conclusion regarding these results, this difference in complications is likely the most important finding of the study. Future investigation with larger patient numbers is necessary to provide evidence-based guidelines and to better define what should be the standard-of-care approach for surgical management of these injuries.
The present study is subject to the limitations and biases inherent to any retrospective investigation with a relatively small overall sample size, despite being larger in size than previous efforts on the same topic. Regardless of the attempt to account for the small sample size in statistical analysis by performing nonparametric tests for continuous numerical data and exact tests for categorical data, the study did not reach an adequate power to make statistically meaningful conclusions. Based on previous research utilizing the Pedi-IKDC, the KOOS-Child, and HSS-Pedi-FABS as outcomes, a priori power analysis determined that 17 patients per group would have been the minimum necessary sample size to detect a difference between operative cohorts with 80% power. This study was limited to cohorts of 10 patients each. There is a possibility of selection bias given that patients in the study were treated by a group of 7 different surgeons. There is also the potential for recall bias regarding the PRO responses, given the wide range of time from surgery to PRO completion. Moreover, PRO responses were not obtained preoperatively, and therefore, baseline preinjury status could not be compared with postoperative status. Lastly, while there were no occurrences of residual angular deformity or leg-length discrepancy seen in this study, observation of a greater number of patients for a longer period may reveal additional information on this topic, given the associated risk of physeal injury with these injuries and procedures. Future investigation with larger numbers of patients is necessary to further support these results and better explore differences in long-term outcomes and complications after these 2 surgical approaches.
Conclusion
Tibial-sided PCL avulsion fractures are rare injuries that can be effectively treated with either open or arthroscopic repair. Most patients healed without problems and returned to high activity levels, regardless of surgical approach. Further prospective investigation with larger numbers of patients is warranted to better support these results, especially the possibility of the open approach allowing for a biomechanically superior fixation construct than that of the arthroscopic technique, given the differences in postoperative complications seen in each cohort.
Footnotes
Final revision submitted February 10, 2025; accepted March 10, 2025.
One or more of the authors has declared the following potential conflict of interest or source of funding: Y-M.Y. is a consultant for Smith & Nephew and OrthoPediatrics. D.K. has received consulting fees from MIACH Orthopaedics, DePuy Synthes Products, Smith & Nephew, and J&J; and education payments from Arthrex Inc and Kairos Surgical. M.K. has received consulting fees and royalties from OrthoPediatrics and has received education payments from Kairos Surgical. B.H. is a consultant for and owns stock in Imagen Technologies Inc; is an educational consultant for and has received education payments from Arthrex Inc, Kairos Surgical, and Pylant; has received consulting fees from DePuy Synthes Products; has received nonconsulting fees from Arthrex Inc; has served on the Board of Directors for the Pediatric Research in Sports Medicine; has served as a committee member and committee chair of Pediatric Orthopaedic Society of North America; and receives royalties from Springer Science & Business Media. AOSSM checks author disclosures against the Open Payments Database (OPD). AOSSM has not conducted an independent investigation on the OPD and disclaims any liability or responsibility relating thereto.
Ethical approval for this study was obtained from Boston Children’s Hospital (IRB-P00040649).
